Analysing the Current State of a Warehouse

-A Framework Based on VSM, Activity Profiling and Benchmarking

Ida Wessman and Maja Bärring

Master Thesis for M.Sc. in Mechanical Engineering

Faculty of Engineering, Department of Industrial Management and Logistics

Supervisor: Joakim Kembro, Lund University

i

Acknowledgements

This thesis is conducted during the spring of 2014 as a final part of our Master’s Degree in Mechanical Engineering at Lund University, Sweden. The thesis has been conducted on behalf of Lund University and Alfa Laval AB.

We would like to thank our supervisor Joakim Kembro at the Department of In-dustrial Management and Logistics, Lund University for valuable comments and feedback during the project. We would also like to thank all employees at Alfa Laval that have been a part of this project. A special thank to Bertil Ljungberg, Lean Manufacturing Specialist at Alfa Laval for the support during the project and the Friday meetings’ attendees for initiating the project.

Lund, May 2014

Ida Wessman Maja Bärring

ii

Abstract

The 21st century has brought significant changes to the market. The ability to de-liver the right product at the right time with the right price is no longer just crucial for a company’s competitive success, in the long-term perspective it will decide if the company will survive on the market at all. A warehouse has a major impact on a firm’s service levels, response times and overall cost. To meet customer re-quirements and to keep up with the ever-changing market there is a need of im-proving warehouse operations.

For this study, a framework with Value Stream Mapping (VSM) in combination with Activity Profiling and Benchmarking is developed. VSM visualises the flow of material and information in a process and it is a concept commonly used for industrial improvements. The use of VSM in warehouses is however lagging be-hind. Activity Profiling is used to understand the activities and operations con-nected to a warehouse and Benchmarking is the process of gathering and sharing assessments of performance. It is believed that the combination will give a more comprehensive result compared to applying only one of the concepts. To meet the purpose of the thesis a case company is used, Alfa Laval in Lund. Alfa Laval has two warehouses at the site that will be used to examine and evaluate the frame-work. The purpose of the study is to investigate how the framework is applicable to present a current situation analysis of a warehouse. The aim of combining the three different concepts is to create a framework that takes several aspects into account in order to establish a fair picture of the warehouse state.

For this study “the research process onion” serves as an outline for the methodolo-gy. Data collection is in the centre of the model and the core of this method. Inter-views, observation and secondary data have been used for the data collection to receive a fair picture of the current state in the two warehouses at Alfa Laval.

With the framework data are analysed and improvement areas are identified. A proposal is given of where Alfa Laval could build a new warehouse at the site. It is then analysed how the three concepts contributed to given recommendations and identified improvements. The three concepts complement each other in many ways and it is recommended to use them together when evaluating the current ware-house state and to find areas of improvement.

Key Words: Value Stream Mapping, Activity Profiling, Benchmarking, Warehous-ing, Warehouse design, Lean warehouse, Distribution centre and Warehouse management

iii

Sammanfattning

Under 2000-talet har betydande förändringar skett på marknaden. Ett företags förmåga att leverera rätt produkt, i rätt tid och till rätt pris är inte längre bara viktigt för företagets framgång utan kommer också vara avgörande för företagets långsiktiga överlevnad på marknaden. Med hjälp av ett lager kan företaget påverka sina servicenivåer, svarstider och kostnader. För att möta kundernas krav och för att hålla jämna steg med den ständigt föränderliga marknaden finns det ett behov hos företag att förbättra sin lagerverksamhet.

Under detta examensarbete har ett ramverk baserat på Value Stream Mapping (VSM), Activity Profiling och Benchmarking utvecklats. De tre koncepten är kombinerade för att ta hänsyn till flera aspekter och för att kunna presentera en verklighetstrogen bild av hur lagerverksamheten fungerar. VSM är ett verktyg som kartlägger flödena av material och information. Konceptet är vanligt förekom-mande för utvärdering av tillverkningsprocesser men användningen av konceptet i logistik- och lagersammanhang är inte lika utbrett. Activity Profiling används för att förstå de operationer och aktiviteter som ingår i en lagerverksamhet. Bench-marking innebär att man samlar in information om sin verksamhet och jämför mot ett idealt värde. Kombinationen av koncepten antas bidra till ett bättre helhetsre-sultat än om bara ett av koncepten hade använts. Syftet är att utvärdera ramverkets lämplighet för att presentera en nulägesanalys av ett lager. För att kunna utvärdera ramverket har det applicerats på Alfa Lavals två lager i Lund.

För denna studie har en forskningsmetodik som kallas för ”the research process onion” använts. Intervjuer, observationer och andrahandsdata har använts som datainsamlingsmetoder för att få en verklighetstrogen bild av lagerverksamheten på Alfa Laval i Lund.

Med hjälp av ramverket analyseras sedan nuläget och förbättringsområden identi-fieras. Ett förslag ges också på var Alfa Laval skulle kunna placera ett nytt lager. Vidare analyseras hur de tre koncepten bidragit till det givna förslaget och till att ge förslag på förbättringar. De tre koncepten kompletterade varandra på ett givan-de sätt under studien och rekommendationen är att de tre koncepten ska användas tillsammans för att få en helhetsbild av en lagerverksamhet.

Sökord: Value Stream Mapping, Activity Profiling, Benchmarking, Warehousing, Warehouse design, Lean warehouse, Distribution centre och Warehouse manage-ment.

iv

Table of Contents

1   Introduction ...................................................................................................... 1  1.1   Background ........................................................................................................ 1  1.2   Purpose and Research Questions ........................................................................ 2  1.3   Description of Case Company ............................................................................ 3  1.4   Delimitations ...................................................................................................... 4  1.5   Structure of Thesis ............................................................................................. 4  

2   Methodology ......................................................................................................... 5  2.1   Introduction ........................................................................................................ 5  2.2   Research Philosophy .......................................................................................... 6  2.3   Research Approach ............................................................................................ 7  2.4   Research Strategy ............................................................................................... 8  

2.4.1   Case Study Design ............................................................................. 9  2.5   Time Horizon ..................................................................................................... 11  2.6   Data Collection Methods .................................................................................... 12  2.7   How the Study was Performed ........................................................................... 13  2.8   Credibility in Research ....................................................................................... 17  

2.8.1   Qualitative and Quantitative Data ..................................................... 17  2.8.2   Reliability .......................................................................................... 18  2.8.3   Validity .............................................................................................. 18  2.8.4   Generalizability ................................................................................. 19  2.8.5   Objectivity ......................................................................................... 19  

3   Frame of Reference .............................................................................................. 20  3.1   Warehouse Operations ....................................................................................... 20  3.2   Explanation of the Framework ........................................................................... 21  3.3   Value Stream Mapping ....................................................................................... 22  

3.3.1   Waste ................................................................................................. 22  3.3.2   Product Family Matrix ....................................................................... 24  3.3.3   Map the Current State ........................................................................ 25  3.3.4   Map the Future State .......................................................................... 26  3.3.5   Why VSM? ........................................................................................ 27  

3.4   Actvity Profiling ................................................................................................. 28  3.4.1   Customer Order Profiling .................................................................. 29  3.4.2   Purchase Order Profiling ................................................................... 30  

v

3.4.3   Item Activity Profiling ....................................................................... 30  3.4.4   Inventory Profile ................................................................................ 31  3.4.5   Calendar-Clock Profile ...................................................................... 31  3.4.6   Activity Relationship Profile ............................................................. 32  3.4.7   Investment Profile .............................................................................. 33  

3.5   Benchmarking .................................................................................................... 33  3.5.1   Warehouse Performance Measures .................................................... 33  3.5.2   The Gap Analysis .............................................................................. 35  

3.6   How to Adapt the Framework ............................................................................ 37  4   Identifying the Current State ................................................................................ 39  

4.1   Introducing the Case ........................................................................................... 39  4.1.1   Identified Problem Areas in the Warehouses .................................... 40  

4.2   VSM ................................................................................................................... 41  4.2.1   Identification of Customers and Suppliers ........................................ 41  4.2.2   Mapping the Current State ................................................................. 43  

4.3   Activity Profiling ................................................................................................ 49  4.3.1   Utilisation .......................................................................................... 56  

4.4   Benchmarking .................................................................................................... 58  4.4.1   In-time Delivery Accuracy ................................................................ 58  4.4.2   Picking Accuracy ............................................................................... 59  4.4.3   Receiving Accuracy and Supplier Quality ........................................ 59  4.4.4   Picking Productivity .......................................................................... 60  4.4.5   Warehouse Cost ................................................................................. 61  

5   Identifying Areas of Improvement ....................................................................... 62  5.1   The Warehouse Performance Gap Analysis ....................................................... 62  5.2   VSM ................................................................................................................... 63  5.3   Activity Profiling ................................................................................................ 65  5.4   Benchmarking .................................................................................................... 67  

5.4.1   World-Class Measures ....................................................................... 69  6   Recommendations ................................................................................................ 70  

6.1   VSM ................................................................................................................... 70  6.2   Activity Profiling ................................................................................................ 71  6.3   Benchmarking .................................................................................................... 72  6.4   The Combination of the Concepts ...................................................................... 73  6.5   A Suggestion for the Future State ...................................................................... 73  6.6   The Gap Analysis for a Future State .................................................................. 76  

7   Analysis of Framework ........................................................................................ 77  

vi

7.1   VSM ................................................................................................................... 77  7.2   Activity Profiling ................................................................................................ 78  7.3   Benchmarking .................................................................................................... 79  7.4   The Framework .................................................................................................. 80  

8   Conclusions .......................................................................................................... 83  8.1   Answering the Research Questions .................................................................... 83  8.2   Future research ................................................................................................... 88  

References ................................................................................................................... 89  Appendix A ................................................................................................................. 92  

 

vii

List of figures

Figure 1 - The research process onion (Saunders et al., 2003: 83) .......................... 5  Figure 2 - The research philosophy is divided into three areas shown in the figure: positivism, realism and interpretivism (Saunders et al., 2003: 83-85) .................... 6  Figure 3 - The three prevailing research approaches within logistics: inductive, deductive and abductive (Kovács and Spens, 2005) ................................................ 7  Figure 4 - Case study design (Voss et al., 2002) ................................................... 10  Figure 5 - Three data collection methods were used during this study and those are: observations, interviews and secondary data (Saunders et al., 2003) ............. 13  Figure 6 - A description of how the study was performed .................................... 15  Figure 7 - The three concepts that will be used in order to evaluate a warehouse and the warehouse operations ................................................................................ 21  Figure 8- A product family matrix (Rother and Shook, 2004: 4) .......................... 24  Figure 9 - Different symbols used when mapping a value stream (Bicheno, 2009: 66-74; Rother and Shook, 2004: 9-34) .................................................................. 25  Figure 10 - An example of VSM (Rother and Shook, 2004: 9-34) ....................... 26  Figure 11- An example of activity relationship profile (Frazelle, 2002a: 42-43). . 32  Figure 12 - An example of a warehouse performance gap analysis (Frazelle, 2002a: 55-58) ......................................................................................................... 35  Figure 13 - How to adapt the framework and what each concept contributes with37  Figure 14 - A PHE and the different products included ........................................ 39  Figure 15 - The figure presents customers and suppliers for warehouse A (Workshop with warehouse A) .............................................................................. 41  Figure 16 - Customers and suppliers for warehouse B are presented (Workshop with warehouse B) ................................................................................................. 42  Figure 17 - Presents the material flows at the site in Lund (Participant observations in the warehouses and at the yard) ......................................................................... 43  Figure 18 - VSM of warehouse A (VSM with warehouse A) ............................... 45  Figure 19 - Placement of activities in warehouse A (VSM with warehouse A) .... 45  Figure 20 - VSM of warehouse B (VSM with warehouse B) ................................ 47  Figure 21 - Shows were different activities are located in warehouse B (VSM with warehouse B) ......................................................................................................... 47  Figure 22 – Item popularity distribution warehouse B .......................................... 49  Figure 23 - Item popularity distribution warehouse B ........................................... 50  

viii

Figure 24 - The different colours describe how often one pallet position in the warehouse is visited ............................................................................................... 51  Figure 25 - Warehouse A: first level ..................................................................... 51  Figure 26 - Warehouse A: second level ................................................................. 52  Figure 27 - Warehouse A: third level .................................................................... 52  Figure 28 - Warehouse B: first level ...................................................................... 53  Figure 29 - Warehouse B: second level ................................................................. 53  Figure 30 - Warehouse B: third level ..................................................................... 53  Figure 31 - Lines per order distribution warehouse A ........................................... 55  Figure 32 - Lines per order distribution warehouse B ........................................... 55  Figure 33 - Suppliers delivering products with poor quality ................................. 60  Figure 34 - Warehouse Performance Gap analysis for warehouse A and B .......... 62  Figure 36 - Steps that should be removed in warehouse B according to the VSM 64  Figure 35 - Steps that should be removed in warehouse A according to the VSM 64  Figure 37 - How to store products warehouse A ................................................... 66  Figure 38 - How to store products warehouse B ................................................... 66  Figure 39 - A future state for the warehouse at Alfa Laval ................................... 74  Figure 40 - Warehouse layout inside the warehouse ............................................. 75  Figure 41 - Warehouse performance gap analysis of the future state .................... 76 Figure 42 - Describes how the framework should be applied ............................... 81 Figure 43 - Wastes identified by each concept and the order that the framework should be performed .............................................................................................. 84 Figure 44 - Which of the concepts that evaluate the different warehouse operations ............................................................................................................................... 85

   

ix

List of tables

Table 1 - Description and purpose of the different research strategies ................... 8  Table 2 - Timeline and persons involved in the process of collecting data ........... 17  Table 3 - The seven wastes (Hines and Rich, 1997; Liker and Meier, 2006: 36; Bicheno, 2009: 20) ................................................................................................. 23  Table 4 - Translation of wastes for a warehouse ................................................... 24  Table 5 - Identified gaps of VSM applications in a warehouse ............................. 27  Table 6 - Customer order profiling (Frazelle, 2002a: 15-44) ................................ 29  Table 7 - Item activity distributions (Frazelle, 2002a: 30-37; Bartholdi and Hackman, 2010: 221) ............................................................................................. 31  Table 8 - Key Performance Indicators (KPI) (Bartholdi and Hackman, 2010: 239) ............................................................................................................................... 33  Table 9 - Performance measures (Frazelle, 2002a: 54-55; Manzini, 2012: 11-12) 34  Table 10 - Frazelle’s model over warehouse operation practices (Frazelle, 2002a: 67) .......................................................................................................................... 36  Table 11 - Information about the different activities and storage areas in warehouse A (VSM with warehouse A) ................................................................ 46  Table 12 - Information about the storage areas and activities in warehouse B (VSM with warehouse B) ...................................................................................... 48  Table 13 - Wastes identified for both warehouses (VSM with warehouse A and VSM with warehouse B) ........................................................................................ 48  Table 14 - Information about the two warehouses related to the Activity Profiling ............................................................................................................................... 49  Table 15 - Order increment distribution for warehouse A ..................................... 54  Table 16 - Order increment distribution for warehouse B ..................................... 54  Table 17 - The activity relationship profile for warehouse A ................................ 56  Table 18 - The activity relationship profile for warehouse B ................................ 56  Table 19 - Presents the total amount of pallet positions in both warehouses and the utilisation ............................................................................................................... 57  Table 20 - Presents the utilisation of plastic bins in the two warehouses .............. 57  Table 21 - Presents the utilisation in the three tents .............................................. 58  Table 22 - Summarises the utilisation for storage in the two warehouses and in the three tents ............................................................................................................... 58  Table 23 - Reasons for delays for warehouse B .................................................... 59  Table 24 - Picking accuracy warehouse B ............................................................. 59  

x

Table 25 - Waiting and overproduction identified for both warehouses ............... 63  Table 26 - An example of a product that is stored in six different locations ......... 67  Table 27 - Reasons for delays to one of warehouse B’s customer ........................ 69  Table 28 - The two warehouses are positioned against world-class ...................... 69      

1

1 Introduction

The introduction includes a background description of the research topic, the pur-pose and the research questions for this thesis. A description of the case company is given and the delimitations for the thesis are presented.

1.1 Background Smaller shipments, faster deliveries, higher quality, increased variety and availa-bility; these factors are all by now taken for granted by the customer (New, 1993; Manzini, 2012: 2). The ability to deliver the right product, at the right time with the right price is no longer just crucial for a company’s competitive success. In the long-term perspective it will decide if the company will survive on the market at all (Christopher and Towill, 2001). The 21st century has brought significant changes to the market - the uncertainty has grown rapidly, the competition has become even more crucial, a product’s life cycle is shortened, the customer de-mand is inconsistent and uncertain and the suppliers are unreliable (Rimiené, 2011). The present situation states the importance of a well-functioning supply chain to meet customer requirements and expectations (Christopher and Towill, 2001). The supply chain is the chain of activities that includes planning, coordinat-ing and controlling the flow of information, material, parts and finished goods from the supplier to the customer (Stevens, 1989).

Frazelle (2002a: 5) argues that the warehouse plays a key role in the success, or failure, of modern supply chains. A warehouse has a major impact on a firm’s service levels, response times and overall costs (Bozer, 2012: xii; Bartholdi and Hackman, 2010: 5). To meet customer requirements there is a need of improved warehouse operations, which involve receiving, put-away, storage, picking, and shipping of goods (Constantino et al., 2012). Despite the stated importance of warehouse operations, they are considered as both costly and time consuming (Bartholdi and Hackman, 2010: 3). For this reason firms are limiting warehouse resources in terms of labour, equipment and space. With many actors and decision makers, warehouses are rather complex things to deal with (Manzini, 2012: viii). The prevailing situation calls for new ways to evaluate warehouses and reduce warehouse costs.

One concept that has been widely used for improvements in the manufacturing sector over the last decades is lean. Lean is today one of the most widely used approaches for industrial improvements; the focus is mainly on the elimination of

2

wastes and costs (Liker and Meier, 2006: 33). In this setting wastes are all activi-ties that do not add any real customer value (Liker and Meier, 2006: 34). Even though this concept has been widely used in the manufacturing sector, its applica-tion in warehouses is lagging behind (Bartholomew, 2008). Dotoli et al. (2013) propose that the lean principles could be used in order to manage a warehouse more efficient. One concept that originates from the lean principles is Value Stream Mapping, abbreviated VSM. VSM visualises the flow of material and in-formation in a process. According to Dharmapriya and Kulatunga (2011) there is a potential to apply VSM to a warehouse. Identified gaps in the literature show there is an opportunity to investigate how VSM is applicable to a warehouse.

For this study, a framework with VSM in combination with two already estab-lished concepts for warehouse investigation, Activity Profiling and Benchmarking, is developed. Activity Profiling is used to understand the activities and operations connected to a warehouse (Bartholdi and Hackman, 2010: 235). In alignment with VSM, Activity Profiling includes both the information and material flow (Frazelle, 2002a: 27). Since VSM only visualises the activities involved and not how they actually operate, Activity Profiling serves the purpose to conduct more detailed information about the different activities involved in the warehouse. Benchmark-ing is the process of gathering and sharing assessments of performance (Hackman et al., 2001; Frazelle, 2002b: 65; Gu et al., 2010) and has been successfully ap-plied to a variety of business functions and industries (Frazelle, 2002b: 65).

1.2 Purpose and Research Questions The purpose of this study is to investigate how a framework constituted by VSM, Activity Profiling and Benchmarking is appropriate to present a current situation analysis of a warehouse. The idea is to combine the benefits of VSM, which can visualise the material flows in the warehouse, with Activity Profiling and Bench-marking that are previously used concepts for examining a warehouse (Frazelle, 2002b: 65, 231-241; Balk and de Koster, 2008). The purpose and aim of combin-ing the three different concepts, is to create a framework that takes several aspects into account in order to establish a fair picture of the current situation in a ware-house. It is believed that the combination gives a more comprehensive result com-pared to applying only one of them. In order to understand how the framework contributes to new knowledge and how it should be used, it is tested and examined on a case company in order to gain real-life experiences of how it operates. The aim is to identify the current state at the case company, find areas of improvement and develop a proposal for how a warehouse could be improved with the help of the framework. The research questions are:

3

RQ1: How do the three concepts complement each other and why should they be used together in order to evaluate a warehouse?

RQ2: How do the three concepts overlap and were there any complications with using them together?

RQ3: Why should VSM be included in the framework and what benefits were found by applying this concept?

1.3 Description of Case Company To answer the research questions and meet the purpose of the thesis, a case com-pany is used to evaluate and examine the framework. The case company selected for this study is Alfa Laval. Alfa Laval has developed products since 1833 and their products are involved in reducing carbon emissions, treating water and min-imising energy consumptions. They launch between 35 and 40 products every year, holds over 1,900 patents and their financial goal is to grow with 12 % yearly. Production is spread on several continents, 32 major manufacturing units, and customers are all over the world. They manufacture products such as heat ex-changers, separators, pumps and valves. Of those plate heat exchangers have the strongest market position with a share of more than 30 %. Priorities for Alfa Laval are safety, quality, delivery and costs (www.alfalaval.com).

One of the 32 manufacturing units is situated in Lund. There are two larger ware-house locations at the site, hereinafter referred to as warehouse A and B. Those locations are used to store raw materials and semi-finished goods. These kinds of facilities are known as production warehouses and cover the vast majority of warehouses (van Den Berg, 1999). Therefore, the manufacturing unit in Lund is an appropriate choice of case for this study. Space in both warehouses are utilised to a high extent and because of the space limitations some components are stored out-side and in temporary tents, resulting in a number of problems. Components are damaged by rust, barcodes detach from the product, sheets get stuck together in the production, and pallets are moist and bring in water and dirt in the production. Going back and forth to the tents contributes also to excessive transportations at the site and extra handling of material. Picking and inspection are performed man-ually and most components are stored in pallet racks. The number of stocked items at the plant increases constantly and the site is in need of an investigation of how the activities are operating and where potential improvements can be identified. It is believed that there are possibilities to utilise the current space in a more profita-ble way. If the space crunch has no solution Alfa Laval is thinking of building a new warehouse. Another objective for Alfa Laval to initiate this project is a desire to release more space in the factories for workshop areas with the long-term goal to complement with new production at the site.

4

1.4 Delimitations Handling of raw material is not included in the project, while a plan already exists of how to store raw material in the future. Goods arrival, goods departure, work in process and the organisation is not either included. Storage, retrieval and restock-ing policies are not included and nor is the Warehouse Management System, WMS. All information flows are excluded from the research due to the time limi-tation of this project.

1.5 Structure of Thesis The structure of the report clarifies what each chapter and section includes. The second chapter, the methodology, explains how the study was accomplished. Dif-ferent approaches and theories connected to research are introduced. The third chapter presents the frame of reference including a comprehensive explanation of the three concepts to be evaluated in this study. The first concept explained is VSM: definition of waste, how to map current and future state. It is important to understand waste in order to interpret the VSM. Activity Profiling is the second concept to be explained. Different profiles are explained that will help to under-stand how e.g. customers place their orders and the dependency between different activities in a warehouse. Benchmarking is the last concept to be explained. The part about Benchmarking presents performance measures used for warehouses and how a warehouse could be evaluated by a gap analysis. The fourth chapter, identi-fying the current state, presents the current situation in the two warehouses identi-fied by the three concepts. In the fifth chapter areas of improvement are presented. Recommendations and a final suggestion are discussed in chapter six. The sugges-tion is based on information gained from the three concepts. Finally the three con-cepts are evaluated and conclusions are presented.

5

2 Methodology

The chapter gives an explanation of different approaches and theories connected to methodology for research studies. The methodology presented serves the pur-pose to guide and explain the chosen approach for this study and give a back-ground for how the study was accomplished.

2.1 Introduction There are numerous methods and theories applicable for research. For this study “the research process onion”, initial introduced by Saunders et al. (2003: 83), serves as an outline. In the centre of the model are the data collection methods, see figure 1.

Research  philosophy  

Research  strategy  

Time  horizon  

Research  approach  

Data  collection  methods  

Figure 1 - The research process onion (Saunders et al., 2003: 83)

6

The chosen research titles for this study are marked bold in the figure. The onion has five layers: research philosophy, research approach, research strategy, time horizon and data collection methods. Every layer will guide the researcher in the right direction, so that the final data set answers the stated research questions. Eve-ry layer will be further explained with the intention to give a comprehensive back-ground for chosen methodology.

2.2 Research Philosophy Before determining research strategy and method, the philosophy has to be settled according to Saunders et al. (2003: 83-85). In this model the philosophy is divided into three areas: positivism, realism and interpretivism, see figure 2. Positivism is comparable to the philosophy of natural science. A positivistic philosophy implies that the observations can serve as basis for generalizable laws applicable on other situations with the same circumstances. The researcher shall not have any impact on the research; neither shall the subject of the research affect the researcher. Con-trary, interpretivism states that generalizable laws cannot describe the social world of business and management as for the natural science. A business situation is far more complex and unique to be explained by a law. A research that has an inter-pretivistic position aims to explain the subjective meanings why people act in a specific way in a situation and the generalizability of the research is not as crucial as for the positivistic position. An additional step in understanding peoples’ moti-vation to act in one way is to adapt the position of realism. With this philosophy, the researcher aims to explain the environment that influences a person’s actions unconsciously. It is social constructions and factors that will be vital for a person’s views and behaviour.

Figure 2 - The research philosophy is divided into three areas shown in the figure: positivism, realism and interpretivism (Saun-ders et al., 2003: 83-85)

Research Philosophy

Positivism

Natural science

Realism

Social science

Interpretivism

Business and management

7

The aim of this study is to investigate and examine how a framework based on three different concepts is applicable for a warehouse at a manufacturing compa-ny. The study gives generalizable suggestions and recommendations on how to use this framework and how each of the concepts contributes. The aim with the study is to investigate and present recommendations and guidelines for how to use this framework for future investigations of a warehouse. The positivistic philosophy is selected, since the results give generalizable laws for future applications.

2.3 Research Approach Three research approaches within logistics are prevailing according to Kovács and Spens (2005). These are the inductive, deductive and abductive approach, see fig-ure 3. Research within logistics has historically mainly applied either a deductive or inductive approach. The deductive approach begins with theory and applies it on the specific case. The inductive approach contrariwise, starts with a specific case to end up with a general law. The abductive approach is a combination of deductive and inductive. It is initiated with a theory, which is applied on the case and ends up with a result. Arbnor and Bjerke (2009: 15-19) declare that abduction starts from fact in the same way as induction, but is anyhow closer to deduction since it does not ignore theoretical knowledge at a start as for induction.

At an initial stage in this study a comprehensive literature review was performed. The purpose was to gather information within the field and find existing gaps in the literature. Based on the literature study, a framework was established. In order to understand how the framework contributes to new knowledge and how it should be used, it is tested and examined on a case company in order to gain real-life ex-periences of how it operates. From this examination conclusions can be drawn and recommendations can be presented. How this study was performed is very much in alignment with how a deductive approach is performed. Therefore, a deductive research approach is chosen.

Deductive

Theory → Case

Inductive

Case → Result

Research Approach

Abductive

Theory → Case → Result

Figure 3 - The three prevailing research approaches within logistics: inductive, deductive and abductive (Kovács and Spens, 2005)

8

2.4 Research Strategy The main objective of a research is to be able to answer the initial stated research questions according to Saunders et al. (2003). A research strategy can be viewed as a general plan for how the data collection should be performed to answer the research questions. The strategy involves decisions about sources for data collec-tion, clear objectives and delimitations for the research. Research strategies in-cluded in the onion are: experiment, survey, case study, grounded theory, ethnog-raphy and action research. An overview of each strategy is provided in table 1.

Table 1 - Description and purpose of the different research strategies

Research strategy Description Purpose Experiment (Denscombe, 2009: 75-77)

Derives from natural science and involves testing and evaluating estab-lished theory.

Identify causes to a phenomenon and observe the influence of special factors on the research object.

Survey (Denscombe, 2009: 25-26;Saunders et al., 2003: 92)

A large set of data is collected from a predetermined target group in a standardised procedure. Enables comparable results.

Measure some aspects of a social phenomenon or trend and test theory.

Case study (Denscombe, 2009: 59-73; Yin, 2009)

In-depth investigation of a situation in its real-life environment using multiple source of evidence.

Understand the complex rela-tions between factors influenc-ing a social setting.

Grounded theory (Denscombe, 2009: 125-147; Saunders et al., 2003: 93)

Aims to explain what really happens in real-life situations instead of test-ing existing theories.

Explore a new subject to provide new insights and theories. Can also clarify concepts.

Ethnography (Saunders et al., 2003: 93)

Study a social subject in its natural environment over an extended time period.

Describe cultural practices and traditions for a social phenome-non.

Action research (Denscombe, 2009: 169-181; Saunders et al. 2003: 93-95)

The researcher takes an active role in the organisation for the study and aims to change it simultaneously.

Solve problems of more practi-cal character and set guidelines for best practice.

Experiment implies that the researcher shall be able to influence and manipulate parameters included in the research. This is not the situation in this study and ex-periment is therefore disregarded as a research strategy. So are survey, grounded theory and ethnography. Survey is neglected since the study does not require statements from a large target group, grounded theory since the study is a deduc-tive research based on theory and ethnography is neglected because the study is time constrained. The question that remains is whether a case study or action re-search shall be used. Important aspects for this study are to participate in the re-search subject to understand how it operates. This would justify an action research. It is however neglected because it requires a cyclic procedure. Planning, imple-

9

mentation, monitoring and evaluation should be done in cycles during the study and that is not an appropriate procedure for this study (Saunders et al., 2003: 95).

Yin (2009: 8-14) explains that the form of the question is an important factor when deciding research strategy. The form refers to who, what, where, how and why. A case study is the most appropriate research strategy to apply when the researcher aims to answer how and why. Since that is the situation in this study, a case study should be the best option. The case study will give a rich understanding of the processes at the site in Lund, which is sought, and various data collection methods can be used. With a case study existing theories can be explored and challenged, which is of interest for this study. It can even provide the researcher with new hypothesis (Saunders et al., 2003: 93). Based on these arguments, a case study is chosen as research strategy.

2.4.1 Case Study Design The procedure of establishing a case study follows a number of steps that can be seen in figure 4. All steps and how the case for this study was designed are ex-plained in this paragraph. The first step is to settle the framework and the research questions for the study. The framework serves the purpose to highlight what is important for the study and forces the researcher to decide at an early stage which aspects and variables that shall be included. It is important to formulate the re-search questions appropriate from the start to guide the collection of data in the right direction (Voss et al., 2002). For this study a framework was established after an extensive literature review. Based on background, problem description, purpose of the study and the framework, the research questions were formulated.

For the next step in designing a case study, the assignment is to choose one or multiple cases and also the type of case (Voss et al., 2002; Yin, 2009: 46-62). In this study one case was chosen: Alfa Laval’s site in Lund. Even though two ware-houses are included in this study they are considered as one case. They have some differences, but are anyhow relatively equal and they will constitute a good base for the Benchmarking. VSM is a well-established concept within production de-velopment and therefore a production unit was searched for this study. The pro-duction unit in Lund suits this precondition very well in order to investigate how the framework is applicable for a production unit. The site has complex inbound material flows and one of the most important customers to the warehouses is the production line, which are circumstances that are desired for this study.

The third step is to develop research instruments and protocols for collecting data. For a case study the primary sources of data are interviews, structured respectively unstructured. To conduct an interview in a desirable way, questions and subjects that shall be covered are formulated in a protocol. The number of respondents

10

shall also be decided; it is a trade-off between efficiency and richness of data (Voss et al., 2002). An initial action in this research was to spend time in the warehouses to gain an understanding of how each organisation is structured. These observations were crucial for the upcoming steps in the study. Another crucial aspect when collecting data was to involve personnel with diverse area of exper-tise and responsibility. Therefore time was devoted to contact the right persons to participate in focus interviews and when mapping the VSMs. Questions were for-mulated in advance and the procedure was identical for both warehouses; the same questions were asked at both occasions. The interview guide is to be found in ap-pendix A. Other sources of data used in this study are explained in the section “Data Collection Methods”.

When conducting the field research, the researcher must decide on whom to con-tact. A principle informant, well informed about the data anticipated, is preferred. The situation may however be that the organisation is not aware about who is to be referred to as the principle informant. In that case it is vital to get in contact with a person that can provide the researcher with access and put the researcher in contact with the right persons (Voss et al., 2002). In this study the supervisor at Alfa Laval was an important person for communicating the right contacts, both at an initial stage and throughout the entire study. As previous stated, interviews were one data

Figure 4 - Case study design (Voss et al., 2002)

Case study design

1. Research framework

and questions

2. Choosing cases

3. Developing reserach

instruments and protocols

4. Conducting field reserach

5. Reliability and validity in case research

6. Data document-ation and coding

7. Analysis

11

collection method and can be performed by a single investigator. To be multiple investigators may however increase the confidence in the findings and also en-hance the creative potential of the team stresses Eisenhardt (1989). Two investiga-tors performed the study, which will increase the confidence in the findings. While collecting data, notes shall be taken and participants shall be given the opportunity to take part of the notes afterwards and give comments (Voss et al., 2002). This was also performed throughout the study. The next step in the procedure of de-signing a case study is “Reliability and validity in case research”. Since this is explained more into depth in the section “Credibility in Research Process”, this is not further explained here.

After collecting data, the set should be documented and coded. The first step is to perform a detailed documentation according to what is defined in the protocol. It can include writing down notes and/or transcription of tapes. It can also imply gathering documents and other material from the field or through other sources. Ideas and insights that arise while conducting data shall also be documented (Voss et al., 2002). For this study no case visits or interviews were recorded, but notes were taken. These notes were discussed and documented afterwards. Ideas and impressions were also included in the documents. To increase the accuracy, partic-ipants at the occasions were given the opportunity to give comments on the docu-ments. While drawing the VSMs a portable white board was used to document notes and comments. Everyone involved could then agree on the accuracy in the comments and that the documentation was done correctly.

For the analysis Eisenhardt (1989) suggests two steps: analysis within case data and searching for cross-case patterns. The first step is to analyse the pattern of data within each case. It is of importance to investigate each case as a stand-alone enti-ty before the researcher tries to generalize the patterns to other cases. The central idea is to compare theory with data repetitively. A close fit between theory and data is significant, as mentioned, to yield an empirically valid theory (Eisenhardt, 1989). Case research can both test stated hypotheses and develop new hypotheses (Voss et al., 2002). For this study a single case was chosen, Alfa Laval’s site in Lund. Therefore, only analysis of a single case can be performed and to search for cross-case patterns is not relevant. The aim is to find patterns and give recommen-dations on how to apply this framework. It will provide generalizable laws for future use.

2.5 Time Horizon It is of significance for a researcher to decide the time horizon of the study. The question to be asked is whether the study should reflect only a short sequence of time, referred to as cross sectional, or if the study is carried out for an extended

12

time period, longitudinal. A cross sectional time horizon is common to adapt, since most studies are time constrained. The longitudinal horizon has the ad-vantage that it will obtain changes and developments over an extended time hori-zon. The cross sectional time horizon is adapted in this study (Saunders et al., 2003:95). The study reflects only a short period of time and it is time constrained.

2.6 Data Collection Methods Data collection is the core subject of the research onion. Secondary data, observa-tions and interviews are data collection methods used in this study. Sampling and questionnaires are also data collection methods included in the onion. The ad-vantage of applying sampling and questionnaire is that the researcher can gather opinions and perceptions from a large group of people in a standardised way and this will enable comparison (Saunders et al., 2003). This is however not crucial for this study and they are therefore not used for data collection.

An observation implies that the researcher observes e.g. while a task is performed. The purpose of an observation is to receive real-life experience, which will support the understanding of how something behaves or appears. An observation can be either a direct observation or a participant observation. For a participant observa-tion the researcher participates in the activities and is not only observing, which is what distinguishes participant observation from direct observation (Yin, 2009: 109-113). For this study initial observations, in accordance with participant obser-vation, were crucial for the understanding of how the warehouses at Alfa Laval operate. These were performed meanwhile the literature study was performed and the research questions were formulated in order to give the study a depth. When it was decided that VSM was one of the concepts in the framework, observations were required once again. A VSM is an observation and investigation of the actu-ality, a direct observation.

Interview is an essential source of information for the case study. There are differ-ent types of interviews depending on purpose: in-depth interview, focused-interview and formal interview. An in-depth interview is not restricted to one sin-gle event and the interviewee is asked about key information and personal opin-ions. A focused-interview is restricted to one occasion and previous gained infor-mation is validated. Therefore it is of importance to not ask leading questions. Focus interview is the most formal interview of them and it provides the research-er with a structured way to ask questions. It provides the researcher with quantita-tive data (Yin, 2009: 106-109). When performing the VSMs it was important that everyone involved were well informed about the procedure of a VSM and what was expected. Therefore two preparatory workshops were held before the VSMs were performed. Since the material and the agenda for the workshops were decid-

13

ed and prepared in advance, the workshops were equivalent to focus interviews. The interview guide and additional information are to be found in appendix A.

Secondary data are already collected data for another purpose that can be reana-lysed for the purpose of the study in order to answer the research questions (Saun-ders et al., 2003: 188). In this study, files with large data sets were requested from different persons at Alfa Laval. The files contained order data for a longer period, in most cases more than one year. Secondary data were mainly used for the Activi-ty Profiling and Benchmarking. The data collection methods applied are summa-rised in figure 5.

2.7 How the Study was Performed In some degree the methodology applied for this study has already been explained, but this section aims to explain it more in detail including the precise procedure for how the research was performed. It is a detailed explanation in order to enable a reconstruction of the study and to give a rich understanding of how the thesis was carried out. At the very first step for this study the scope of the project was decided. It was determined together with personnel working at Alfa Laval: Bertil Ljungberg, Bengt Larsson, Håkan Nilsson, Jens Richter and Tobias Augustsson. These persons are considered to have a broad overview of the factory and the warehouses as well as knowledge of which areas that are in need of investigation. Together with these persons the project and the delimitations for the study were set. Every second week a meeting together with this group of people was held to make sure that the thesis was moving in the right direction. Bertil Ljungberg was the supervisor at Alfa Laval and he introduced a number of people that would be helpful for the study; people familiar with Alfa Laval’s information system and personnel in the warehouse organisations. These contacts were helpful during the collection of secondary data and when performing the mapping.

Data collection methods used in this study

Observations

Direct observations during the VSMs

Participant observa-tions for the initial study

Secondary data

Secondary data mainly used for Ac-tivity Profiling and Benchmarking

Interviews

Focus interviews held during the workshops and VSMs

Figure 5 - Three data collection methods were used during this study and those are: observations, interviews and secondary data (Saunders et al., 2003)

14

To receive a deeper understanding of the current situation in the warehouses with respect to material flows and handling, time was dedicated to meet personnel working in the warehouses and to take part of the daily operations, participant observations. Two days were spent in warehouse A and one day in warehouse B. Experiences gained from these visits in the warehouses and the factory were help-ful to comprehend the problem scenario. Based on experiences gained from the visits a map was done visualising the flows of material within and between the warehouses. Team Manager for warehouse B validated the map and he is consid-ered to have profound knowledge about the material flows over the whole plant. He was asked to comment the data, to add missing information and to correct any errors.

The literature study was an essential part from the start of the project and was commenced in an early stage. For the literature study Google scholar and Lovisa (lub.lu.se) served as Internet search engines. Published papers have been the pri-mary literature source because they are peer reviewed. Peer reviewed means that one or more persons review the article with equal competence within the field before publishing. Key words while searching have been Value Stream Mapping, Warehouse Design, Lean Warehouse, Distribution Centre, Activity Profiling, Benchmarking and Warehouse Management. Books have been used as a supple-ment for compiling the methodology chapter and the frame of reference. For the frame of reference a number of books have been used: World-Class Warehousing and Material Handling and Supply Chain Strategy written by Edward H. Frazelle. They have mainly complemented the sections about Activity Profiling and Benchmarking. Frazelle is a president and CEO of Logistics Recourse Internation-al and he is also a founder of The Logistics Institute at Georgia Institute of Tech-nology (Frazelle, 2002b: 358). Frazelle’s books are considered as reliable sources and have been recommended by the supervisor at Lund University. For the section about VSM a book named Learning to See: Value Stream Mapping to Add Value and Eliminate MUDA written by Mike Rother and John Shook has been an im-portant information source. Warehouse & Distribution Science written by John J. Bartholdi and Steven T. Hackman has also served as a complementary source of information for the frame of reference. This book is used as course literature at Lund University and therefore considered as a reliable source.

From the literature review some gaps could be identified, presented in the intro-duction, and this has been a motivation for the selected concepts constituting the framework. The first gap identified was that lean principles and VSM could be used in warehouses in a wider extent. The second gap was that methods that could evaluate how a warehouse operates and how the activities in a warehouse depend on each other were needed. A decisive part of the study was also to formulate ap-propriate research questions, abbreviated RQs. The literature study in combination with knowledge gained by visits in the factory served the purpose to finally formu-

15

Step 1. Defining the project scope

Step 2. Visits to the warehouse

Step 3. Literature study 1

Step 4. Identified gaps

Step 5. Literature study 2

Step 6. Develop the framework

Step 7. Formulate the RQs

 

Step 8. Applying the framework          

Step 9. Analyse the framework

 

Step 10. Conclusions  

• Performing the VSMs • Collecting secondary data • Performing the Activity Profiling and Benchmarking • Identify areas of improvement based on the framework • Present recommendations and a suggestion for the future state

Figure 6 - A description of how the study was performed

late the RQs and they were vital for the result. The RQs were also discussed to-gether with the supervisor at Lund University and at Alfa Laval. The task formu-lated by Alfa Laval has been to evaluate the current situation in the warehouses and also to give recommendations whether a new warehouse would be necessary or if they could handle all the components by just doing some changes of the current situation. These thoughts will be answered and a suggestion will be given.

So far the information presented has mainly concerned the preparatory work for the study. As mentioned, participant observations were done in order to gain an understanding for how the site in Lund functions. Therefore the next step in the study was to collect the data required for the empirical part of the study. This in-volved performing the VSMs and hosting the workshops. Workshops were held with the purpose to inform about the project and to define focus areas. Customers and suppliers to the warehouse were identified as well as their expectations and requirements. This in order to make sure that everyone had the same idea about VSM and what the goal with the project was. After the information was given some questions were asked, presented in appendix A. The same questions were asked to both groups. The attendees defined what they thought were important for their own processes. At the end of the workshops a brain storming method called affinity diagram was used to identify problems in the warehouses. The purpose was to find problem areas to focus on during the study. The attendees wrote their thoughts on post-its. They were allowed to write anything they could think of as a problem, at least three things. All thoughts were then discussed and the post-its were divided into groups. One group of post-its contained thoughts that mainly meant the same thing. The workshop and the answers given were compiled as well as the affinity diagrams. The compilations were sent to all attendees for validation.

16

Team Manager and Unit Manager attended the workshop for warehouse A. A se-cond workshop was held for people working in warehouse B. Team Leader Re-ceiving and Shipping, Team Manager and Unit Manager attended that workshop. Team Leader for warehouse B was also asked to attend the workshop but was not available.

Subsequently the next step when collecting data was to perform the actual VSMs. Two separate occasions, one for each of the warehouses, were scheduled. Ques-tions for every step in the mapping were prepared in protocols, securing that noth-ing was forgotten while mapping and also that the same questions would be asked at the two different occasions. The protocol can be found in appendix A. A porta-ble white board was brought to the warehouses together with some additional equipment necessary for the mapping. The VSMs were done in groups and each occasion lasted for about three hours. At the first occasion Team Leader, Team Leader Receiving and Shipping and an employee in the warehouse were participat-ing in completing the VSM in warehouse A and to answer the questions. The warehouse employee is working with inspection in warehouse A and he is well familiar with the material flows since he has been working in the warehouse for a longer period. Team Manager and Team Leader Receiving and Shipping helped to complete the VSM in warehouse B. Since it was not feasible to map all the exist-ing material flows, focus was on the material flows from the largest customers for both warehouses. The personnel provided the team with this information. Some flows could also be mapped as one group because the personnel handled them almost in the same way. The mapping started at the customer and the material flows were followed counter current. Personnel involved and the time frame for the study are summarised in table 2.

After conducting the current state map, information was needed in order to com-plete the maps and in order to present some additional findings about the ware-house functions. For this purpose, secondary data were requested from the person-nel at Alfa Laval. Data were mainly necessary for accomplishing the Activity Pro-filing and Benchmarking. Employees handling financial questions and supporting the information system at Alfa Laval were contacted. They helped by gathering order data and other information concerning the warehouses and the products stored. Data were also collected through email by some of the persons mentioned in this section.

When the secondary data needed for the next step were collected, Activity Profil-ing and Benchmarking were initiated. For the Activity Profiling it involved inter-preting the order data files and presenting the findings in an appropriate way. Benchmarking was completed by gathering information about each warehouse that can be compared against each other and against what is considered as world-class standards. The results from this step will eventually be analysed and help to an-swer the RQs. Finally the analysis and conclusions remained as the last steps in

17

this study. The analysis was completed based on findings done during the empiri-cal study. Conclusions were made and the RQs were answered.

Table 2 - Timeline and persons involved in the process of collecting data

Steps in the data collection Time frame Persons involved Defining the project scope and delimitations

Dec. 2013 - Jan. 2014

Bertil Ljunberg - Specialist of Lean Manufactur-ing Bengt Larsson - Senior Manufacturing Manager Håkan Nilsson - Production Manager Jens Richter - General Manager for Plate Heat Exchangers Tobias Augustsson - Senior Specialist of Lean Manufacturing

Participant observations in the warehouses and at the yard

Jan. 2014 Team Leader warehouse A Team Manager warehouse B

Workshop with warehouse A 2014-02-18 Team Manager warehouse A Unit Manager warehouse A

Workshop with warehouse B 2014-02-12 Team Manager warehouse B Unit Manager warehouse B Team Leader Receiving and Shipping

VSM with warehouse A 2014-02-20 Team Manager warehouse A Team Leader for Manufacturing Frames Compo-nent Team Leader for Receiving and Shipping

VSM with warehouse B 2014-02-25 Team Manager warehouse B Team Leader for Receiving and Shipping

Collecting secondary data Mar. 2014 – Apr. 2014

Employees handling financial questions and sup-porting the information system at Alfa Laval

Visits to the warehouses May 2014 Personnel in the warehouses

2.8 Credibility in Research When gathering data for the empirical study, there are some aspects that are im-portant to take into account to ensure the credibility in the research. In this chapter qualitative respectively quantitative data, reliability, validity, generalizability and objectivity in research are discussed.

2.8.1 Qualitative and Quantitative Data There are some distinctions between qualitative and quantitative data. Most studies involve some numerical data that need to be quantified before analysing it. This is referred to as quantitative data and it can imply e.g. the frequency of occurrences or more complex data. The collection results in numerical and standardised data that can be presented with the use of diagrams and statistics (Saunders et al., 2003: 378). In this study quantitative data are used, mainly for the Activity Profiling.

18

Large data files are gathered and examined with the aim to present data for the analysis of how the warehouses are organised. Qualitative data are collected as well. Qualitative data are based on meanings expressed through words and the collection is not standardised as for quantitative data. Since it is not standardised, the results need to be categorised (Saunders et al., 2003: 378). In this study quali-tative data are gathered mainly at the workshops and when mapping the VSMs. Warehouse personnel are asked about their opinions and experiences, their an-swers are then categorised into different groups.

2.8.2 Reliability Saunders et al. (2003) define three conditions to be crucial for the study to fulfil the requirements of reliability. First of all, the study has to be reproducible at an-other occasion and give the same result. Secondly, another observer shall obtain the same observations as the researcher. Thirdly, there should be a sense in how conclusions were drawn from the conducted data. If all these aspects are fulfilled, the study can be seen as reliable. The first condition is fulfilled since information about how the study was accomplished is recorded and documented throughout the study. This will ensure the reproducibility of the study. The second condition can be seen as fulfilled as well, since the report has been peer reviewed both by other master thesis students and by the supervisor at Lund University. Also the last condition is fulfilled with the same arguments as for condition two.

2.8.3 Validity The purpose of validity is to make sure that the results really measure what was intended to be measured. If the data set is able to answer the research questions is in this matter significant. Validity concerns also the accuracy in the data collec-tion. By adapting triangulation validity can be increased (Denscombe, 2009: 184-190). Depending on intention with the study, different combinations can be estab-lished with two or more sources: data triangulation, investigators triangulation, methodological triangulation, theoretical triangulation or analytical triangulation according to Thurmond (2001). Data source triangulation means that data are col-lected from different sources. Data should not be collected from one single person and not be restricted to one occasion or one place. In this study many visits were done to the warehouses and different persons took part in the interviews. Investi-gators triangulation means that more than two researchers should take part in in-terviews and observations. Discussion and confirmation of data leads to greater credibility. This requirement is fulfilled since there are two researcher involved in this study. To achieve methodological triangulation multiple data collections methods should be used. In this case observations, interviews and secondary data

19

are used and therefore the requirement for methodological triangulation is achieved. Theoretical triangulation may refer to the procedure of testing and ana-lysing the same data set with different theories. This is done in this study when comparing the three concepts VSM, Activity Profiling and Benchmarking. Analyt-ical triangulation is when two or more methods are used to analyse data. Accord-ing to Thurmond (2001) analytical triangulation could be to validate the collected data. Validation of data is done in all cases in this study.

2.8.4 Generalizability When conducting a study, it will be questionable if the results have any meaning for future research. Saunders et al. (2003: 102-103) refer to this matter as a re-search’s generalizability. If the findings are applicable for other cases or organisa-tions the research can be considered as generalizable. The aim of this study is to evaluate how a framework combining VSM, Activity Profiling and Benchmarking is applicable for the warehouses in this study. This is the primary aim, but another expected result is to investigate whether this framework is applicable for other warehouses with similar conditions as in this case. A desired outcome from this study is that the results shall be generalizable for future research. In order to in-crease the generalizability in this research a positivistic research philosophy has been chosen. This philosophy will likely give generalizable results. In qualitative research the participants shall be selected purposefully for the contribution they can make to the research that will also contribute to the generalizability (Morse, 1999).

2.8.5 Objectivity To ensure that the researcher’s subjective thoughts are excluded and not have any impact on the result, it is important to investigate the objectivity in the research states Denscombe (2009: 379). An objective research is not biased and the collect-ed data can be seen as neutral and accurate for the study. To obtain objectivity Roberts et al. (2006) say that the researcher should try to achieve an analytical distance to the research and that the researcher should be non-reactive to the re-search.

20

3 Frame of Reference

The frame of reference outlines the theoretical foundation that is used as a refer-ence and guide throughout the research process. The chapter is introduced with an explanation of warehouse operations with the aim to give the reader a basic understanding for this topic. Thereafter VSM, Activity Profiling and Benchmark-ing are presented and explained.

3.1 Warehouse Operations The most common operations in a warehouse are receiving, put-away, storage, picking and shipping. At the receiving department the arriving products are noti-fied and inspected. Damages, incorrect quantities and other errors are noted. Re-ceived products are registered so that they are known to be available and payments are dispatched (Bartholdi and Hackman, 2010: 22). Tasks performed at the receiv-ing department have an impact on the sequencing activities in the warehouse states Frazelle (2002a: 74). Incorrect received products will complicate put-away, stor-age, picking and shipping. Before put-away, a storage location must be determined for the product. Storage locations that are available, how much weight they can bear and how large they are must be known at all times. The storage locations should be scanned to record where the products are placed (Manzini, 2012: 2). For the issue of storing, there are two aspects to take into account. On the one hand, pallet storage system should be chosen with the aim to maximise volume utilisa-tion and density in the warehouse. On the other hand, the pallet retrieval system should be designed as beneficial as possible for the picking activities. These two aspects stand in contradiction to each other and both of them are of high priority for a world-class warehouse (Frazelle, 2002a: 85-108).

Picking operations are very cost intensive activities and the topic has been investi-gated in numerous papers. There are different ways to perform this operation: sin-gle order picking, batch picking, zone picking, pallet picking and free-form pick-ing. Single order picking is when orders are completed one at a time. Batch pick-ing is when orders are batched and items are picked from pallet positions as they are visited by the picker. Zone picking is when a picker is dedicated to a picking zone. Workload has to be balanced among the zones. When zones are automatical-ly adjusted to keep balance this is named a bucket brigade and when an order is passed from zone to zone this is named a pick and pass system. In free-form pick-ing, the picker is free to pick from any location (Manzini, 2012: 4-5). Further it is

21

possible to do broken case picking, case picking or pallet picking. An example of broken case picking would be if 12 bolts would be picked out of a cartoon contain-ing 50 bolts. Case picking is when 50 bolts would be picked, which is the whole cartoon. An example of pallet picking is when 1000 bolts would be picked and a pallet contains 20 cases, which is the whole pallet. After the picking operation products are packed, loaded into freight carriers and transported to the customer. Those are the activities included in the shipping operation. Bartholdi and Hackman (2010) explain warehouse operations in their book and it is recommended for fur-ther reading.

3.2 Explanation of the Framework In this study a framework based on the three concepts VSM, Activity Profiling and Benchmarking is developed. The basis for the framework can be seen in figure 7. The purpose of the framework is to enable an extensive investigation and evalua-tion of the current situation in a warehouse. Each concept will be further explained in this chapter.

The three concepts will provide information about how the warehouse operations perform and where potential improvements are to be found. The understanding of how the warehouse operates and the dependency between activities is important to be able to improve the warehouse.

Activity Profiling

Bench-marking VSM

Receiving Put-away Storage Picking Shipping

Figure 7 - The three concepts that will be used in order to evaluate a warehouse and the warehouse operations

22

3.3 Value Stream Mapping Lean principles have successfully been applied in the manufacturing industry dur-ing the last decades in order to reduce costs and increase quality. Value Stream Mapping (VSM) is a concept within the lean principles that visualises material and information flows. It can be used to identify where value is added to the product and when wastes occur (Rother and Shook, 2004: 2). The aim of VSM is to mini-mise the resources consumed at every activity and to identify which activities that do not add any real value. Real value implies all activities that add real customer value state Serrano et al. (2008). VSM is a paper and pencil method that includes both the information and material flow along a products way in the value stream. A VSM has the benefit that it can visualise the entire material and information flow, from the moment the raw material arrives until it is ready for shipping (Rother and Shook, 2004: 2). Along the flow of material and information wastes are identified and also the cause for each waste. A VSM provides the team with a common language that facilitates the discussion and decision-making concerning a future state.

The warehouse includes many different material and information flows. The flow of information can go from the customers through an information system to the personnel working in the warehouse. The material flow goes through the different warehouse operations: receiving, put-away, storage, picking and finally shipping (Bartholdi and Hackman, 2010: 22).

3.3.1 Waste Waste was first presented 1988 in the Toyota Production System by Taiichi Ohno (Abdi et al., 2006). Ohno presented seven wastes: transportation, inventory, mo-tion, waiting, over processing, overproduction and defects. Each waste is further explained in table 3. Overproduction, to produce more or earlier than the customer wants, is considered to be the worst waste because it contributes also to the other wastes such as inventory and transportation (Liker and Meier, 2006: 36). More recently, other kinds of wastes have been suggested. One of them is the waste of people, which means not letting people do what they are good at or not make use of their ideas (Bicheno, 2009: 20; Liker and Meier, 2006: 36). To spend time on urgent matters instead of important ones has also been suggested as a waste (Su-zaki, 1987: 8). There are activities that are necessary despite the fact that they are wasteful though. It can be distances that cannot be eliminated in order to pick up a part or the need of transferring a tool from one hand to another. These kinds of activities may not be possible to eliminate (Hines and Rich, 1997).

23

Table 3 - The seven wastes (Hines and Rich, 1997; Liker and Meier, 2006: 36; Bicheno, 2009: 20)

Table 3 explains the wastes that can be found in a manufacturing unit. These wastes could however be translated to wastes in a warehouse and these definitions are given in table 4. Consequences and the ideal state are the same for a warehouse as in table 3.

Waste Definition Consequences Ideal state Overproduction To produce more or earli-

er than requested from the customer.

Excessive lead and storage times. Large WIP-stocks and detection of defects are impeded.

To produce what the customers want when they want it.

Waiting Appears whenever goods are not moving or being processed.

Time is not used efficient, which affects both workers and goods.

No waiting time with a consequent faster flow of goods.

Transportation Any movement in a pro-duction plant that is non-value adding.

Reduced quality due to distance of communication. A fixed cost is generated.

Minimisation of transportation rather than elimina-tion.

Over pro-cessing

When too complex solu-tions are found for a simple task. Over pro-cessing can also imply unnecessary machining of a part.

A product is produced with higher quality than required.

Manufacture ac-cording to the expected and re-quired quality.

Inventory Inventory in itself is a waste, since it does not add any real customer value.

Increased lead-time and space. Prevents rapid identi-fication of problems, since they are hidden by the in-ventory.

Reduced inventory levels.

Motion Implies ergonomic issues for the employees. To stretch, bend and pick up when these actions could be avoided.

Tiring for the employees. Can lead to poor productivi-ty and quality problems.

Unnecessary mo-tions are eliminat-ed.

Defects Products with faults. Defects result in either discard or reprocessing, both costly for the company.

Zero defects.

24

3.3.2 Product Family Matrix It is not convenient to map all the products passing through the facility when com-pleting a VSM. The customer might as well be more interested in one specific product and therefore it is advisable to choose a product family. A product family is a group of products that pass through the same processing steps. To identify the product families, a product family matrix is used. It is a table with the processing steps described in the first row and products described in the first column, as can be seen in figure 8.

Table 4 - Translation of wastes for a warehouse

Waste Definition Overproduction To pick, load and ship an order before the customer actually needs it. A custom-

er is in this context the customer to the warehouse, internal at the site or external. Waiting Wherever material and items have to wait before they reach the main storage and

have received a pallet position as well as after. It can be at a temporary storage area or at an activity.

Transportation Transportation in a warehouse is the same as the definition in table 3. Over pro-cessing

Over processing is e.g. to inspect arriving materials and items when there is no reason for doing so.

Inventory To store material and items that does not necessarily have to be stored or are out-of-date. Occupying pallet positions that could be released.

Motion Motion in a warehouse is when operations cause excessive movement for the personnel.

Defects Defects for warehousing are orders that are delivered to the customer with poor quality, at the wrong time, with the wrong quantitative etc. Defects for a ware-house are also all orders that are delivered to the warehouse with poor quality, at the wrong time, with the wrong quantity.

Process steps and equipment

Com

pone

nts

A B C

a x x

b x x

c x x

Figure 8 - A product family matrix (Rother and Shook, 2004: 4)

25

The components with similar processing steps constitute a product family. In the table two products, b and c, belong to the same product family. This is because they pass through the same processing steps, i.e. A and B. Generally the product family with the greatest opportunity for improvement is analysed (Rother and Shook, 2004: 4). For warehousing this could be applied by identifying the material and items that pass through the same operations and follow the same route. The material and items that follow the same material flow from arrival to the customer will constitute the same product family. Instead of process steps and equipment it will say warehouse operations and components are material and items stored in the warehouse.

3.3.3 Map the Current State The first step when creating a VSM is to decide the boundaries of the process. Firstly the product family matrix is performed to identify the product families that will be mapped. After identifying the product families, a current state map should be created which should serve as a baseline for the future state map (Bicheno, 2009: 66-74; Lasa et al., 2008). When creating the current state map, the persons involved should collect the information. Secondary data are not reliable for this matter, neither from a person or a data system. Information should be collected at circumstances that are relevant for the present situation. The mapping start at the customer and the flow is followed counter current. The mapping shall start at the customer because information concerning customer requirements could then be gathered for each step before getting there. The whole material flow has to be viewed in actuality and the fact that every day does not look the same should be ignored. Each step in the process should be presented and contain information such as how many workers that are needed, number of picking errors, working hours and operator expertise.

Suppliers and cus-tomers to the ware-house

Storage

Activity

Transportation

Figure 9 - Different symbols used when mapping a value stream (Bicheno, 2009: 66-74; Rother and Shook, 2004: 9-34)

Different symbols are used to map the process. Those symbols can be seen in fig-ure 9 (Bicheno, 2009: 66-74; Rother and Shook, 2004: 9-34). Both internal and external customers and suppliers to the warehouse are marked with the same sym-

26

bol when mapping. The symbol for activity is used for e.g. inspection, put-away or picking. The symbol for storage is used both when material and items are stored in the main storage and when they are standing waiting on an arriving or departure square. Transportation is marked with an arrow and the map should also include information about the transportations such as distance. Important for the mapping is that everyone involved is familiar with VSM and why it is done. This information could be given during a workshop before the mapping. When the current state is mapped it is possible to identify where in the warehouse wastes could be reduced. Supportive information could be gathered afterwards. By conducting the VSM possibilities for improvements will be identi-fied (Bicheno, 2009: 66-74). Figure 10 shows how a VSM typically look like. Inventories are marked with a triangle and activties with a square. Transportations are marked with an arrow. In figure 10 the flow of information is also included, but it will not be mapped in this study. At the bottom line of the figure is an example of how the map can visualise time spent on each activity or storage area.

3.3.4 Map the Future State The goal with VSM is to eliminate as much waste as possible and all non-value adding activities. In the future state, the processes shall only produce what is need-ed in the next stage. One way of managing this is by applying TAKT. The TAKT time should be the same as the sell rate. A TAKT time of two minutes means that every two-minute shall an order departure from the warehouse to the customer that requests it. In the future state the flow of material should be continuous. A contin-uous flow means that an order or a product is directly moved to the next step with-out any storage between. Where continuous flow cannot be implemented a pull-system should be used instead. In a pull-system each activity only demands an item from the preceding activity by the rate the succeeding activity is consuming

Figure 10 - An example of VSM (Rother and Shook, 2004: 9-34)

27

items (So and Pinault, 1988). Material and items are only stored in the warehouse and not at number of different arriving or departure areas, which would enable a continuous flow of material in the warehouse.

3.3.5 Why VSM? Based on Benchmarking, Hackman et al., (2001) make two conclusions about warehouse performance. The first is that warehouses with a high operating effi-ciency reduce their work content by eliminating material handling steps and by minimising travel time. The second conclusion states that the design quality of material and information flows is the only single consistent indicator of high productivity. The purpose with VSM is to visualise and improve the design of the material and information flow, which is said to be the only single consistent indi-cator of high productivity. Material handling steps can be eliminated and travel distances can be visualised and reduced by a VSM. Table 5 - Identified gaps of VSM applications in a warehouse

Year Identified gaps of VSM applications in a warehouse Author 2003 “Value stream mapping can be a valuable tool for developing and imple-

menting warehousing lean improvement projects.” Garcia, F.C.

2003 “Although many traditional lean techniques maybe difficult to apply, the concepts of improving material flow and eliminating waste can be used to make significant improvement in warehouse lead time.”

Garcia, F.C.

2013 “Although the ultimate goal of lean thinking is to reduce the reliance of manufacturing enterprises on warehousing, yet warehouses are necessary to limit lead times and balanced business processes connections. Conse-quently, there is a need to manage warehouses efficiently. To accomplish this goal, lean principles must be brought into the warehouse.”

Dotoli et al.

2012 “Although companies are involved in the lean change in their entirety with all their different areas, a decisive role is played by internal logistics and particularly warehouse management, which is aimed at optimizing the flow of material and information within the company to maximize profit (Christopher 2010). The related literature includes several application examples of VSM to discrete manufacturing systems (see for instance Keil et al. 2011, McManus and Millard, MOSIM’12 - June 06-08, 2012 - Bor-deaux – France 2002, Ramesh et al. 2008), but few papers employ VSM for lean warehousing.”

Dotoli et al.

2011 “The present records imply that the cost of warehouse operations is com-paratively high due to the existence of many non-value added activities. Therefore, the requirement arises to lean the warehouse operation in terms of cost and time by eliminating non value adding activities and optimizing value adding activities. A Value Stream Mapping (VSM) is one of the best elements in Lean since it shows the collection of all actions (value added & non-value added) that are required to bring a product through the main flows, starting with raw material and ending with the customer (Abdulmalek and Rajgopal, 2007; Rother and Shook, 1999).”

Dharmapriya and Kulatunga

28

Although these similarities seem so apparent only a few papers (Dotoli et al., 2012) have been published that adapt VSM to a warehouse. Based on a literature survey, Gunasekaran et al. (2001) make an attempt to develop a framework for measuring the strategic, tactical and operational level of performance in a supply chain. The study leads to a conclusion that research is needed to streamline the flow of material, simplify the decision-making procedure and eliminate non-value adding activities. These statements do also perfectly match the purpose of a VSM: to simplify decision-making, identify and eliminate non-value adding activities and streamline flow of material and information. A warehouse is a key aspect of modern supply chain states Frazelle (2002a) and therefore it could be a good start to apply VSM in a warehouse. There are gaps of using lean and VSM in ware-houses. Table 5, summarises gaps that are found in literature.

3.4 Actvity Profiling It may appear complicated to understand the dependency between different activi-ties and operations connected to a warehouse. The Activity Profiling could help to simplify the picture. The Activity Profiling supports to describe the activities in a warehouse and is an essential part for the understanding of what really matters in a warehouse. The profiling serves as a baseline to identify problems connected to information and material handling. In the end the Activity Profiling will justify investments and enable required improvements (Bartholdi and Hackman, 2010: 217). Lewczuk and Zak (2013) explain that Activity Profiling involves the analy-sis of historical data, product characteristics and locations, packing patterns and warehouse layout. All these factors are important when identifying potential im-provements.

According to Okeudo and Uche (2013) Activity Profiling can be of great help when analysing activities for the purpose of determining process workflow and layout options. Bartholdi and Hackman (2010: 217) emphasise that it is important to be aware of the customer order patterns and how these affects the workload within the facility, this in order to retrofit and improve the warehouse. Further-more they stress that there are three main types of data that are required to support the profiling: data pertaining to each stock keeping unit (SKU), data pertaining to customer orders, and data pertaining to locations within the warehouse. Basic sta-tistics that have to be identified for a warehouse are (Bartholdi and Hackman, 2010: 217-218):

• Area of warehouse. • Types of storage and material-handling equipment. • Average number of pick-lines shipped per day and the average number of

units (pieces, cases, pallets) per pick-line.

29

• Average number of shipments received in a day. • Average number of customer orders shipped in a day. • Number of order-pickers and shifts devoted to pallet movement, to case

picking, and to broken-case picking. • Average rate of introduction of new SKUs.

All these aspects have an impact on how the warehouse should be organised and planned in order to reach the most desired state of effectiveness. Activity Profiling is divided into several areas: customer order profiling, purchase order profiling, item activity profiling, inventory profile, calendar-clock profile, activity relation-ship profile and investment profile (Frazelle, 2002a: 15-44). Each of them will be further explained.

3.4.1 Customer Order Profiling The purpose of material and information flow through a warehouse is basically to serve the customer’s needs, explains Frazelle (2002a: 15-44). Therefore it is of importance to understand how a customer orders goods to plan and design the operations within a warehouse.

Table 6 - Customer order profiling (Frazelle, 2002a: 15-44)

Customer Order Profiling

Definition Impact on the Warehouse

Family Mix Distribution

Determines whether an order requires items from multiple item families or single item family.

The warehouse can be divided into segments and zones based on item families.

Full/Partial Pallet Mix Distribution

Determines whether case and pallet picking should be sepa-rated.

The warehouse can be divided into different areas.

Full/Broken Case Mix Distribution

Determines whether full and broken case picking should be separated.

The warehouse can be divided into different areas.

Order Increment Distribution

Determines the portion of a unit load requested on a cus-tomer order.

Pre-packing items in quantities that the cus-tomers are likely to order.

Lines per Order Distribution Single line orders

Each order includes one line. Orders can be batched together and create efficient picking tours and zones in the ware-house. Can also imply a forward picking area.

Lines per Order Distribution Ten or more lines per orders

One order includes several lines, more than ten.

Enough work within the order itself so it will represent an efficient work set. Alternatively it can be divided into several orders with a pick and pass system.

Lines and Cube per Order Distri-bution

Defines lines and space re-quired for one order.

Provides crucial information about appropri-ate picking strategy.

30

It is the customer’s orders that drive the system, say Bartholdi and Hackman (2010: 217-225) and the customer order profile can serve as a valuable input for establishing the other profiles included in Activity Profiling (Park, 2012). The different types of customer order profiles are explained in table 6 according to Frazelle (2002a: 15-44). How the customer orders products has an impact on the expected work in a warehouse. Each order is a shopping list containing a single or a number of pick lines. Each pick line generates travel to the item location and subsequently packing, checking and shipping to the customer. Pick lines are there-fore a strong indicator on the work associated in a warehouse. The distribution of picks by order-size, small or large, also indicates which orders that require most aggregate work (Bartholdi and Hackman, 2010: 224-226). For the issue of design-ing a warehouse the customer order profiling is the most important profile to con-sider (Lewzcuk and Zak, 2013).

3.4.2 Purchase Order Profiling The same distributions are included in purchase order profiling as for customer order profiling. The big difference however is that purchase order profiling is in-bound to the warehouse and customer order profiling is outbound from the ware-house. It can be seen as the customer order profiling in reverse (Frazelle, 2002a: 29-30).

3.4.3 Item Activity Profiling The main assignment of the item activity profiling is to designate for each item where it should be located in the warehouse, how much space it requires for stor-age and to what storage mode it should be located (Frazelle, 2002a: 30-37). It will also profile popularity and volume for an individual item (Park, 2012). Another factor identified by the item activity profiling is which items that tend to be or-dered together and this information is important for the decision of storage loca-tion in the warehouse (de Koster et al., 2007). The different types of item activity profiling are clarified in table 7 (Frazelle, 2002a: 30-37; Bartholdi and Hackman, 2010: 221).

To summarise the item activity profiling, it is important in order to know what handling that is required by a specific item and what storage location it should be designated to. Handling characteristics involve physical characteristics such as shape and weight, but also environmental characteristics such as frozen or flam-mable (Park, 2012).

31

Table 7 - Item activity distributions (Frazelle, 2002a: 30-37; Bartholdi and Hackman, 2010: 221)

Item Activity Distributions

Definition Impact on the Warehouse

Item Popularity Distribution

A minority of the items in a warehouse generate a majority of the picking activi-ties.

Assign the most popular items to the most accessible warehouse locations.

Cube-Movement Distribution

To assign items to storage modes based on their cube-movement.

Depending on cube size, different storage systems can be used.

Popularity-Cube-Movement Distribution

Both item popularity and cube-movement are considered.

The distribution considers both popularity and cube-movement.

Item-Order Completion Distribution

Identifies small groups of items that can fill a large group of orders.

The small groups of items that are frequently used can be organised in one zone in the warehouse.

Demand Correlation Distribution

Items tend to be requested together and this method serves to identify the correla-tion between individual items and families of items.

Place items that are correlated nearby each other.

Demand Variability Distribution

Indicates the daily standard deviation for each item based on the daily demand.

By identifying the standard devia-tion the aim is to prevent the need of restocking during a day.

3.4.4 Inventory Profile The inventory profile includes two distributions: item-family inventory distribu-tion and handling unit inventory distribution (Frazelle, 2002a: 38). For a ware-house with a lack of storage possibilities, the source of the problem may be con-nected to the inventory rather than the actual lack of space in the warehouse. The item-family inventory distribution aims to identify which products that are con-sumed in a faster rate than other items. Items that are stored for a longer period should be evaluated whether it is necessary to store them at all. An ABC analysis is applicable for this situation. The ABC analysis classifies products based on ac-tivity. A is the small fraction that account for most of the activity, B the medium fraction that account for a medium part of the activity and C the bulk of products that account for the small part of the activity. Bartholdi and Hackman (2010: 218) are just two authors explaining the ABC classification and the topic has been dis-cussed in many papers.

3.4.5 Calendar-Clock Profile The purpose with a calendar-clock profile is to identify peaks and valleys in ware-house activities. Based on this information, a material handling system can be sized appropriate and scheduling of staff can be established. The calendar-clock

32

profile has two distributions: seasonality and daily activities distribution. Frazelle (2002a: 40-41) clarifies that both of them indicate peaks and valleys in inventory as well as for receiving, shipping and returns, but on different time horizons.

3.4.6 Activity Relationship Profile An activity relationship profile reveals the internal relations between activities and functions in a warehouse. It makes suggestions on how blocks of activities should be organised relative to one another in a block layout. The purpose with this ma-trix is to recognise which activities those are dependent on each other along the material and information flow (Frazelle, 2002a: 42-43). de Koster et al. (2007) have also discussed this topic and its impact on warehouse layouts. The reason for importance is to highlight how different activities depend on each other and how that affects the localisation of activities in a warehouse.

The abbreviations in the example, figure 11, represent the dependency and the importance of localising activities nearby. The number declares the reason for importance and the letter the proximity importance. There are in total nine differ-ent reasons for importance, but in this study only number seven and three are used. Number seven is “shared space” and number three “material flow”. The letter ex-plains the proximity importance as mentioned. A stands for absolutely necessary, E for especially important, I for important, O for ordinary important, U for unim-portant and X says undesirable (Frazelle, 2002a: 42-43). As can be seen in the example, figure 11, it says 7E for inspection and receiving. The result, 7E, says that receiving and inspection should be localised near each other and that they should share space in a future solution in order to enable a desirable material flow.

ACTIVITY R

EC

EIV

ING

INSP

EC

TIO

N

PUT

-AW

AY

PIC

KIN

G

LO

AD

ING

RECEIVING - 7E 3E 3I 3O

INSPECTION 7E - 7E 3I 3O

PUT-AWAY 3E 7E - 7E 3I

PICKING 3I 3I 7E - 3I

LOADING 3O 3O 3I 3I -

Figure 11- An example of activity relationship profile (Frazelle, 2002a: 42-43).

33

3.4.7 Investment Profile To make appropriate suggestions on design and investment decisions, an invest-ment profile can serve the purpose to indicate costs and operating parameters nec-essary for the decision-making, explains Frazelle (2002a: 42-44).

3.5 Benchmarking In accordance with any other enterprise, a warehouse should constantly measure its performance. The performance measures should be compared against other companies and the organisation should plan on how to improve these (Bartholdi and Hackman, 2010: 239). Benchmarking is the process of gathering and sharing assessments of performance and it can result in a plan for improvements (Hack-man et al., 2001; Frazelle, 2002a: 46-69; Gu et al., 2010). An assessment is based on a number of management ratios and performance indicators, in table 8 some of them are explained (Balk and de Koster, 2008; Bartholdi and Hackman, 2010: 239-248).

Table 8 - Key Performance Indicators (KPI) (Bartholdi and Hackman, 2010: 239)

KPI Example Operating cost The warehouse costs as a percentage of sales. Operating productivity Typically pick-lines, orders, or pallets handled per person-hour. Cycle time Measured by dock-to-stock time or warehouse order cycle time, expressed

by minutes per order. Order accuracy Fraction of shipments with returns.

According to Frazelle (2002a: 46), there exist three perspectives of benchmarking: internal, external and competitive. Internal benchmarking is benchmarking within the firm. External benchmarking looks outside the firm’s industry and competitive benchmarking compares firms in the same industry. In all cases, benchmarking is done to identify strengths and weaknesses and to learn from the best according to the assessment. Bartholdi and Hackman (2010: 239) stress that an attempt to com-pare the existing warehouse to the ideal warehouse generally does not generate much suggestions and furthermore does not give any hint on how to reach that state. More advisable is to compare warehouses with comparable circumstances. Benchmarking has been successfully applied to a variety of business functions and industries (Frazelle, 2002b: 65).

3.5.1 Warehouse Performance Measures To measure warehouse performance the same indicators as for business are used. The indicators are financial, quality, productivity and cycle time (Frazelle, 2002a:

34

52-55). Operating cost is the most commonly used financial measure for ware-houses. It is typically measured as the total cost for the warehouse as percentage of sales. Hackman et al. (2001) have anyhow an opposite opinion and say that operat-ing cost should not be used for warehouse assessments. This because operating costs varies directly with product pricing and sales volume that is outside the con-trol of warehouse management. This statement is in agreement with Gunasekaran et al. (2001) arguing that day-to-day controls of manufacturing and distribution operations are better handled with non-financial measures. Also Balk and de Koster (2008) assert that warehouse managers should not influence financial per-formance. In this context they also mention inventory turnover and service levels as variables that are determined at the company decision level and not by ware-house managers. According to Frazelle (2002a: 52-55) financial indicators such as costs for each of the warehouse activities could be used anyhow both for measur-ing improvements and as a base for budgeting.

Productivity is another way to measure warehouse performance. Productivity is defined as the ratio of output resources to the input required for completing that output. Productivity could typically be measured as labour needed for each line picked (Frazelle, 2002a: 54). The output is not necessarily the number of lines picked but can also be weight, number of units handled or number of orders. Likewise labour is not the only input to consider; space and capital investments in equipment are also significant.

Performance measures in a warehouse are put-away accuracy, picking accuracy, inventory accuracy and shipping accuracy, see table 9 (Frazelle, 2002a: 54-55; Manzini, 2012: 11-12). These are performance measures directly relevant to the customers and they are increasingly important (Manzini, 2012: 11). It is possible to compare these kinds of performance measures with world-class firms. In Japan the highest shipping accuracy is 99.997 % and in the United States 99.97 %. This is world-class and this is where to aim for (Frazelle, 2002a: 55-58). Table 9 - Performance measures (Frazelle, 2002a: 54-55; Manzini, 2012: 11-12)

Performance measure Definition Put-away accuracy The percent of items put-away correctly. Picking accuracy The percent of order lines picked without errors. Inventory accuracy The percent of warehouse locations without inventory discrepancies. Shipping accuracy The percent of order lines shipped without errors.

Another performance measure directly relevant to the customer according to Man-zini (2012: 11-12) is the cycle time. Warehouse cycle time can be expressed as dock-to-stock time or warehouse order cycle time. Dock-to-stock is the time from when a receipt arrives until it is ready for picking and warehouse order cycle time is the time from when an order is released to the warehouse floor until it is picked

35

or ready for shipping (Frazelle, 2002a: 55). Another indicator of interest when measuring performance in a warehouse is the utilisation of space, percentage of existing storage location that is actually occupied. It is a fine line between an over-crowded warehouse and underutilised facility. Neither of them is desirable and storage density should therefore be within a world-class range according to Fra-zelle (2002a: 61-62).

3.5.2 The Gap Analysis The gap analysis is used to identify how the existing warehouse performs in com-parison to world-class or the ideal state. The gaps are visually evaluated for a number of measures, three or more.

The measures are placed in a circular pattern. Spokes are drawn from the perfor-mance measures to the centre of the circle. Each spoke is divided with increments of equal value with the best value in the end. The ideal state or world-class is marked on each spoke and each point of measure is joined up and the current state is marked up on each spoke and joined up as well. The outer ring will always de-fine world-class or the ideal state and the inner ring will define the current state, see figure 12. A third ring may represent the goals of reengineering projects, not exemplified in figure 12 (Frazelle, 2002a: 55-58).

Figure 12 - An example of a warehouse performance gap analysis (Frazelle, 2002a: 55-58)

36

Because of the gap analysis’ simple layout it is easy to identify weaknesses and strengths in the existing warehouse. The biggest gap for the most critical measure should be identified. Project goals for the organisation are formulated with the mission to improve the current state based on the gap analysis.

Table 10 - Frazelle’s model over warehouse operation practices (Frazelle, 2002a: 67)

Process Stage 1 Stage 2 Stage 3 Stage 4 Stage 5 Receiving Unload,

stage and in-check

Immediate put-away to reserve

Immediate put-away to primary

Cross-docking Pre-receiving

Put-away First-come-first-serve

Batched by zone

Batched and sequenced

Location-to-stocker

Automated put-away

Reserve Storage Floor storage

Conventional racking bins

Some double deep storage

Some narrow aisle storage

Optimal hybrid stor-age

Picking Pick-to-single-order

Batch picking Zone picking (progressive assembly)

Zone picking (downstream sorting)

Dynamic picking

Slotting Random Popularity based

Popularity and cube based

Popularity cube and correlation based

Pick from reserve storage

Replenishment As needed (pick face complete)

As needed (downstream complete)

Anticipated (by sight)

Anticipated (automated)

Pick-to-trailer

Shipping Check, stage and load

Stage and load Direct load Automated loading

Standards used for continuous feedback

Work Measurement

No stand-ards

Standards used for plan-ning

Standards used for evaluation

Standards used for in-centive pay

Virtual displays

Communications Paper Barcode scan-ning

RF terminals Hands-free Visual displays

Based on these identified factors, investments can be recommended to decrease the distance to the ideal state with consideration to eventual payback time. Fra-zelle (2002a: 67) has developed a tool to evaluate warehouse operation practices based on the gap analysis named the warehouse performance gap analysis. He also says that the warehouse performs as a function of its practices and that practices are what separate world-class performers from the rest. The definitions are shown in table 10. For each activity the different stages towards world-class are ex-plained. Stage five represents world-class standards.

37

3.6 How to Adapt the Framework For this study, a framework based on VSM, Activity Profiling and Benchmarking has been developed and each concept has been given an extensive explanation in this chapter. This section aims to give a preview of how the framework is intended to be applied for this study. Figure 13 shows in which order the framework is to be performed and an explanation of what, why and how is also declared for each con-cept.

Each concept has its pros and cons. The VSM has the benefit that it can visualise the entire material flow from arrival at the site to departure to the customer. It will also provide the team with a common language for the discussions about how the material moves at the site and it is simple to use since it is a paper and pencil method. However the VSM will not provide the same result at every occasion because it is a snapshot of the actuality. The Activity Profiling enables an under-standing for what really matter in the warehouse and it clarifies how the activities

What: visualises the material flow from arrival to departure at the site. Why: to identify wastes and map the physical material flow. How: a paper and pencil method mapped together with warehouse personnel.

What: supports to describe the activities involved in a ware-house. Why: to identify how the activities relate to each other and oper-ate. How: an immense data set is analysed by each profile.

What: the process of gathering and sharing assessments of per-formance. Why: to accomplish a comparison against other warehouses. How: identify ware-house performance measures and per-form a gap analysis.

VSM

Activity Profiling

Benchmarking

Figure 13 - How to adapt the framework and what each concept contributes with

38

operate and depend on each other. Disadvantages with Activity Profiling are that it does not visualise the material flow and that a large data set has to be analysed, which can be both time consuming and complex. The Benchmarking will position the warehouse against other warehouses and give measureable goals. It compares the existing warehouse against the ideal state and identifies the gap to ideal state. The drawback with Benchmarking is that every project is unique that hampers the comparison and cannot address problems that not previously have been recog-nised.

39

4 Identifying the Current State

This chapter presents the current state of the two warehouses. First some basic facts about both warehouses are presented, followed by findings from the three concepts VSM, Activity Profiling and Benchmarking.

4.1 Introducing the Case At the site in Lund there are two locations for main storage. They are here referred to as warehouse A and B. Warehouse A stores frame plates, rods and screws among others. Frame plates are heavy and bulky to handle. Warehouse B stores gaskets and plates. Figure 14 shows a plate heat exchanger, PHE, and how it is constructed. The different products that are stored in the warehouses are also illus-trated in figure 14.

Except for the warehouses there are three complementary tents available for stor-age of material that cannot be stored in the warehouses because of lack of space. The tents are referred to as tent 1, 2 and 3. Tent 1 and 2 belong to warehouse B and tent 3 belongs to warehouse A. Products that do not fit in neither the tents nor in the warehouses are stored outdoors at the yard. The two warehouses deliver both to external and internal customers. Internally the two warehouses deliver to different manufacturing units and externally they deliver to different Alfa Laval warehouses that store spare parts and also to other companies that will use the PHEs for their processes. There are no restrictions on how the other sites should

Rod

Plate

Frame Plate

Gasket

Figure 14 - A PHE and the different products included

40

place their orders and spare part warehouses often order large quantities of various products and the demand is therefore unpredictable. The two warehouses receive products from both internal and external suppliers. Internal suppliers are manufac-turing units within the site.

4.1.1 Identified Problem Areas in the Warehouses During the workshops personnel were asked to highlight current problem areas in the warehouses. This was asked with the purpose to become aware of what areas to focus on when performing the VSM, Activity Profiling and Benchmarking.

To enable traceability of products in warehouse A every batch has its own charge number. In this way a product’s origin can be traced when failures occur. If two products are from the same charge this means that they are manufactured in the same batch and that they have the same charge number. Each charge has its own certificate to ensure the traceability and this may reduce the utilisation in the warehouse. If three identical products have different charge numbers this means that they are stored in three different pallets, even though the products would fit in only one pallet. Other circumstances that have to be taken into account for ware-house A, is that products can arrive to the site in Lund before necessary certificates do. Therefore products cannot be released to customers even though they physical-ly are available in the warehouse and occupies pallet positions.

Many problems in warehouse A are related to the suppliers, e.g. mislabelled prod-ucts, products that need to be repacked, incorrect deliveries and pallets that are not charge clean. Other problems concern drawings that are difficult to read and mis-matches between drawings and products. Mismatches could be due to either prod-uct defects or drawings that are out-dated. Inspection is considered as a problem area because of two reasons: it has to be performed partially outside and there is no routine to record information about the suppliers causing the most problems. Because of irregular workload there are problems with planning. This is partly due to irregularly placed orders from the manufacturing units. Because of the difficul-ties in planning the warehouse complete work half-day in advance in order avoid delayed deliveries. When picking smaller products, counting is problematic. Prod-ucts are stored outside which leads to bleached colours and that products are dam-aged by dirt and rust. Products that once are stored in the warehouses are hard to discard. It is however easy to make the decision to store them in the warehouse (Workshop with warehouse A).

Problem areas were also identified for warehouse B. Likewise for warehouse B many of the problems concern suppliers, e.g. that receiving goods are frequently delivered in the wrong quantity. Due to the problems caused by the suppliers in-ventories are high and the workload is irregular because some suppliers only de-liver on certain weekdays. There is much transportation back and forth to the tents and the distance between warehouse B and the tents is long. It is not uncommon

41

that transportations back and forth with the forklifts actually are empty as a result of bad communications. Much time is spent searching for products because of the absence of barcodes and scanners in warehouse B, which also affects the picking accuracy. The manual handling when picking is a problem causing a lot of the picking errors with wrong quantities. These errors lead to significant profit losses and several costly actions such as restoring, shipment and returns management (Workshop with warehouse B).

4.2 VSM This section will present how the VSMs were performed and how the current state was identified in both warehouses with the VSMs. Firstly existing customers and suppliers to the tents are explained and thereafter the actually mapping. Additional information about the snapshot of waiting and overproduced orders are also sum-marised and wastes identified are presented.

4.2.1 Identification of Customers and Suppliers Customers and suppliers, both internal and external, were identified for the ware-houses during an initial workshop. This was important for the subsequent map-ping, since the material flow to the largest customer was of most interest when mapping. Figure 15 presents suppliers and customers to warehouse A (Workshop with warehouse A) and figure 16 presents suppliers and customers to warehouse B (Workshop with warehouse B).

For warehouse A global customers are grouped into one unit because the material flow for all these customers is the same and the only thing that differs is the pack-

Figure 15 - The figure presents customers and suppliers for warehouse A (Workshop with warehouse A)

42

aging. The internal customers are the manufacturing and assembling units and imply that warehouse A is delivering to other departments within the plant. Ware-house A delivers to two manufacturing units, the manufacturing unit of small PHEs and the manufacturing unit of medium sized PHEs with TAKT (the line). Suppliers identified for warehouse A are external suppliers, warehouse B and the department that blasts and paints the frame plates, manufacturing of frame plates. The department for manufacturing of frame plates places the frames in pallet racks themselves, either in warehouse A or at the yard. From warehouse B to warehouse A about 20 orders are delivered every day. Most of the orders, about 90 %, are stored in warehouse A again. External suppliers are considered as one group since products from external suppliers are all treated in the same way (Workshop with warehouse A).

In figure 16 suppliers and customers, both internal and external, for warehouse B are presented. Ronneby is an external manufacturing and sales unit within Alfa Laval. Warehouse B has a number of gasket providers but all material is handled in the same way and therefore they are grouped into one unit. The same applies to providers of plates. The subcontractors are handling among others water jetting and washing. The external suppliers can deliver ten days before the required date. Some of the suppliers are only delivering on certain weekdays to the warehouse.

Figure 16 - Customers and suppliers for warehouse B are presented (Workshop with warehouse B)

43

An internal supplier is the manufacturing unit that manufactures plates; the same unit is also a customer to the warehouse. Warehouse B delivers to warehouse A. All global customers are grouped because the material flow is the same for all of them and the only thing that differs is the packaging (Workshop with warehouse B).

Value for both warehouses is when material and items are delivered correctly packed, at the right time with the right quantity and quality, and with the correct labelling. The arriving material shall also contain information about part number and quantity. Values for their customers are mostly the same factors. Material delivered correctly packed, at the right time with the right quantity and quality, and with the correct labelling (Workshop with warehouse A and Workshop with warehouse B).

4.2.2 Mapping the Current State Figure 17 shows the current state at the site in Lund. The arrows show the material flows. Material arrives to the site and is unloaded either at the dock or out at the yard, illustrated with the two cars in the figure. From there material is transported to either warehouse A, warehouse B or to a storage area at the yard.

There are two buildings at the site and warehouse A is placed in the left building and warehouse B in the right. Material is transported both within the two buildings and between the buildings, as figure 17 shows. Flows of material that are connect-ed to the two warehouses are included in the figure and all others are excluded. The figure shows the different departments at the site. Most of them are either suppliers or customers to the warehouses, but there is also a workshop area that

The Tents

Warehouse A

Storage Yard

Manufacturing Small and Medi-um Sized PHEs

Workshop

Manufacturing large PHEs

Warehouse B

Manufacturing Plates

Figure 17 - Presents the material flows at the site in Lund (Participant observations in the warehouses and at the yard)

44

not earlier has been described. The workshop provides tools required for manufac-turing (Participant observations in the warehouses and at the yard).

Figure 17 gives an overview of the material flows at the site in Lund, but it de-scribes only how the material is transported between the different departments. The VSMs are presented in figures 18 and 20. Those two figures describe the ma-terial flows within the two warehouses. Figure 18 describes the VSM for ware-house A and figure 19 illustrates where it is in the warehouse. At stage 1, material and items arrive outside the warehouse and are subsequently transported into a gateway, stage 2. From the gateway material and items are transported to an in-spection area represented by number 3 in the figure. Number 4 represents put-away, which imply that material and items are taken from the inspection area and are placed in pallet racks. Number 5 in the figure represents material and items stored in pallet racks. Number 6 represents the picking operation. The picked order can take four different ways. What way the order goes depends on the end custom-er. The four end customers are also illustrated in figure 18. Flow A represents the material flow to the line, the triangle demonstrates that the order is waiting before the customer uses it. The B flow represents the material flow to the global custom-ers. Warehouse A has two different activities for global customers: kitting and the spike. Kitting is picking and packing of orders that the customers assemble them-selves and the spike ships to all other global customers. The two activities, the spike and kitting are merged to one flow in the VSM. Material and items going to the global customers are first packed, 7B, and then transported to the gateway, 8B. From the gateway orders are transported to a departure square outside the ware-house, 9B. The C flow represents material flow to the manufacturing unit for small and medium sized PHEs and the D flow represents material flow to the manufac-turing unit for large PHEs. Material and items going to the manufacturing unit for large PHEs are transported between the two buildings while material going to the manufacturing unit for small and medium sized PHEs could be transported within the building. Therefore the flows look different. Figure 19 also shows how the different activities and storage areas are placed in the warehouse (VSM with warehouse A).

Table 11 contains information about the different activities and storage areas for warehouse A. The table contains a snapshot picture of how many orders that were waiting at the moment the mapping was performed, information about how many pallets that could be stored at the different areas, for how long time the orders are waiting at the different areas and the size of the different areas. Number 7C that represents manufacturing of small and medium sized PHEs, has seven different storage areas, each of them presented in the table (VSM with warehouse A).

45

Figure 18 - VSM of warehouse A (VSM with warehouse A)

Figure 19 - Placement of activities in warehouse A (VSM with warehouse A)

46

Table 11 - Information about the different activities and storage areas in warehouse A (VSM with warehouse A)

Location Snapshot (orders)

Pallets possible to store

Waiting time Area (m2)

Arrival square 0 100 2 h 6 x 19 Departure square 0 100 2 h 6 x 19 Gateway 20 60 2 h 5 x 15 Inspection 17 60 1.5 h 13 x 11 Man. (7C):1 6 9 1 day 3 x 1.2 Man. (7C):2 2 24 1 h 6 x 1.2 Man. (7C):3 14 9 2 days 2 x 1.2 Man. (7C): 4 4 4 Pull-system 3.2 x 1.2 Man. (7C):5 8 21 2 days 6 x 1.2 Man. (7C):6 19 5 shelves× 7 m 3 days 1.2 x 8.5 Man. (7C):7 7 30 4 h 11 x 1.2 TAKT (7A) 4 6 TAKT: 20

min 6 x 1.2

For warehouse B most material and items arrive from external suppliers. This is represented by stage 1, 2 and 3A in figures 20 and 21. When material and items arrive from external suppliers they are first placed outside the warehouse and are then transported to the gateway where they are inspected. Material and items arriv-ing from internal suppliers are represented by stage 3B. Material and items are then placed randomly at the floor somewhere in warehouse B. Stage 4 represents put-away, where material are taken from the floor or from the gateway and placed in pallet racks. Number 5 represents material and items that are stored in pallet racks, the main storage. Number 6 represents the picking operations. Depending on the end customer, picked orders have different flows. The material flow going to external customers are represented by 7A and 8A. Orders are placed on a con-veyor where they are inspected and packed, 7A. From the conveyor products are transported to a departure square outside the building. Orders are also transported to internal customers represented by 7B and 8B. The orders are transported to the departure square outside the warehouse, 7B. Orders are then transported to the end customer where triangle 8B is implying that orders are waiting before the custom-er uses them (VSM with warehouse B).

Information about the different activities and storage areas in warehouse B are presented in table 12. The table contains a snapshot picture of how many orders that were waiting at the moment the mapping was performed, information about how many pallets that could be stored at the different areas, for how long time the orders are waiting at the different areas and the size of the area. The floor storage is material and items placed at the floor between the pallet racks. This is represent-ed by number 3B in figure 21. Material and items delivered from internal suppliers

47

are placed at the floor before they are stored in pallet racks, main storage (VSM with warehouse B).

Figure 20 - VSM of warehouse B (VSM with warehouse B)

Figure 21 - Shows were different activities are located in warehouse B (VSM with warehouse B)

48

Table 12 - Information about the storage areas and activities in warehouse B (VSM with warehouse B)

Location Snapshot (orders) Pallets possible to store Waiting time Area (m2)

Floor storage 10 20 7 h 2 x 13 Departure square 4 160 30 min 16 x 10 Arrival square 0 160 1 h 16 x 10

Gateway 0 0 0 h 7.4 x 24.3 Conveyor 6 20 1 h 18 x 8

In table 13 data are presented about the seven wastes. The wastes presented in the table are identified during the mapping of the VSMs. In this case overproduction is orders that are picked but not directly transported and used by the customer. All orders waiting for put-away, transportation, inspection or manufacturing are con-sidered as waiting orders, i.e. all snapshot orders. The average waiting time and transportation distance for an order is presented in the table. How long the distance is depends on the end customer. A lot of time is spent on inspection in both ware-houses even though it may not be necessary; inspection is therefore identified to be over processing. Each product is stored for an average of 54 days in warehouse A and 90 days in warehouse B. At the inspection and kitting unnecessary motion was identified because of missing tables. Personnel do not have the correct equip-ment to carry out the activity ergonomically. No defects could be identified with the VSM in the two warehouses (VSM with warehouse A and VSM with ware-house B).

Table 13 - Wastes identified for both warehouses (VSM with warehouse A and VSM with warehouse B)

Waste Warehouse A Warehouse B Overproduction 64 orders 10 orders Number orders waiting

103 orders 20 orders

Average waiting time

8 h per order 4 h per order

Transportation Material are transported between 240 and 520 m

Material are transported between 490 and 1,030 m

Over processing Inspection by 3-4 persons 7 h a day Inspection by 2-3 persons 7 h a day

Inventory 54 days 90 days Motion Missing tables at inspection and

kitting Missing tables at inspection and kitting

Defects - -

49

4.3 Activity Profiling This section presents data collected according to the principles of Activity Profil-ing in order to present the current situation in both warehouses. Table 14 shows information about the two warehouses related to the Activity Profiling. The table presents the area of the warehouse and the tents. It tells what equipment that are used and the storage type. A default position means that a specific product always is stored in the same pallet position. Most of the picking is broken case picking in both warehouses.

Table 14 - Information about the two warehouses related to the Activity Profiling

Category Warehouse A Warehouse B

Area of the warehouse (m2) 2,611 1,392 Ceiling height (m) 8 8 Area of the tents (m2) Tent 3: 493 Tent 1: 301.4, Tent 2: 828 Height of the tents (m) Tent 3: 6 Tent 1: 5, Tent 2: 7 Material handling equipment Forklifts and barcode system Forklifts Storage type Default position Default position Broken case, full case or pallets picks Broken cases Broken cases Average order lines shipped per day 124 367 Average items per order line 6.5 53.6 Average shipments received in a day (internal orders)

18 95

Average shipments received in a day (external orders)

37 75

Average lines shipped in a day 1160 367 Average lines per order 9.4 3.9 Average rate of introduction of new SKUs

2 per month 2 per month

Working days in a year 254 348

An item popularity distribution showed that a small number of products responded to a large amount of picks, se figure 22 and 23. This relation is more apparent for warehouse B than for warehouse A.

0  

1000  

2000  

3000  

4000  

5000  

6000  

7000  

Num

ber  lines  

Ar-cles  

50

In a warehouse the most frequently picked products should be placed in the most convenient places. Figures 25, 26, 27, 28, 29 and 30 show how often picks are made from different locations in the two warehouses with the help of a bird’s eye. The figures show the three bottom levels, from above, of warehouse A and B. The first level is material and items stored at the floor. The second level is the first shelve and the third level the second shelve. The three lowest levels are presented because those are the most convenient levels to pick from and of most interest to

Figure 22 - Item popularity distribution warehouse A

Figure 23 - Item popularity distribution warehouse B

51

examine. In total there are eight levels of shelves in both warehouses. Those shelves are also examined but they are not visited as frequently as the three bottom levels. Therefore level four to eight are not presented in this thesis. Already when examining the three bottom levels a trend could be seen: the two lowest levels are visited more often than the third level in both warehouses. The black colour indi-cates most visits, the darkest grey second most visited places and the white colour indicates no visits, see figure 24. The data used to create the bird’s eyes are not complete; personnel have visited some places that are marked white during the period and manufacturing orders are probably missing in the analysed order data (Visits to the warehouses).

5264 - 1000 999 - 100 99 - 25 24 - 10 9 - 1 0 940 - 400 399-100 99-11 10-3 2-1 0

A: B:

Figure 24 - The different colours describe how often one pallet position in the warehouse is visited

Figure 25 - Warehouse A: first level

52

Figure 26 - Warehouse A: second level

Figure 27 - Warehouse A: third level

53

Next thing to present is the order increment distribution. Material and items often ordered in a specific quantity in warehouse A are presented in table 15. The two last columns show one or two of the preferred case quantities and a percentage of how many times this case quantity will be useful. This means the percentage of lines no cases will be broken.

Figure 28 - Warehouse B: first level

Figure 29 - Warehouse B: second level

Figure 30 - Warehouse B: third level

54

Table 15 - Order increment distribution for warehouse A

In warehouse B the correlations are not as clear as in warehouse A. There are two articles that could be delivered in cases according to the order increment distribu-tion, see table 16. Besides from the cases defined by the order increment distribu-tion there could be other reasons to pack in case quantities. This could e.g. be when a product often is ordered in large quantities in order to simplify the count-ing.

Table 16 - Order increment distribution for warehouse B

No. of picks Description Package

Size Preferred Case

Quantity No. of times the cases are

useful (%)

290 T5-M Clip-on Peak 300 50 55.9

226 M10 Sheet 700 4 94.6

The lines per order distribution are presented in the two figures, 31 and 32, for warehouse A and B. From figure 31 the conclusion can be drawn that in ware-house A 13,141 orders out of 31,378 orders consist of single lines. This would suggest that orders could be batched together in a larger extent. At the same time 10,755 orders consist of more than ten lines per order, indicating that a large amount of the orders imply enough work in themselves.

For warehouse B most of the orders are single line orders, 14,898 orders out of 25,476 orders. It is also clear from figure 32 that most of the orders are single line orders in warehouse B.

No. of picks Description Package

size Preferred Case

Quantity No. of times the cases are

useful (%) 5634 M16x90 15000 8(4) 96.8(97.4) 2699 M16x140 100 8(4) 93.9(98.4) 1783 M20 700 6 97.6 2258 M20x350 25 6 97.1 1225 M20x200 25 6 98.3

55

To make recommendations on how to localise activities in the warehouses it is essential to identify how the activities depend on each other. Activity relationship profile is therefore performed. Most activities in the two warehouses depend on each other due to the material flow, represented by number three, and a few be-cause of shared space, represented by number seven. The characters are E that stands for especially important, I for important, O for ordinary closeness and U for unimportant. The result for warehouse A is presented in table 17 and the results for warehouse B is presented in table18.

Figure 31 - Lines per order distribution warehouse A

Figure 32 - Lines per order distribution warehouse B

56

Table 17 - The activity relationship profile for warehouse A

Table 18 - The activity relationship profile for warehouse B

Activity

Unl

oadi

ng

Insp

ectio

n

Put-

away

Pick

ing

Hol

e pu

nchi

ng

Insp

ectio

n an

d

pack

agin

g

Loa

ding

Unloading - 3E 3I 3O 3O 3O 3O Inspection 3E - 3E 3E 3O 3O 3O Put-away 3I 3E - 3E 3I 3O 3O Picking 3O 3E 3E - 7E 3I 3I Hole punching 3O 3O 3I 7E - 3O 3I

Inspection and packaging 3O 3O 3O 3I 3O - 3E

Loading 3O 3O 3O 3I 3I 3E -

4.3.1 Utilisation The utilisation was first calculated for warehouse A based on data from an infor-mation system used at Alfa Laval. This data set was only accessible for warehouse A. The warehouse utilisation was calculated to 51.36 %. This number is however not consistent with the real picture. This is because of two reasons: some articles require more than one location in reality but in the system only one location is occupied and some pallet positions available in the system do not actually exist. After consolidation with the warehouse personnel, it was concluded that the data set was not representative and in order to gain accurate data, pallet positions and occupied positions had to be counted manually in both warehouses and in the tents. The results are presented in table 19, 20, 21 and 22. Table 19 presents the total amount of pallets positions in both warehouses and the utilisation. Table 20

Activity

Unl

oadi

ng

Insp

ectio

n

Put-

away

Pick

ing

The

line

The

spik

e

Kitt

ing

Loa

ding

Unloading - 3E 3I 3O 3O 3O 3O 3O Inspection 3E - 3E 3I 3I 3I 3I 3O Put-away 3I 3E - 3E 3E 3E 3E 3I Picking 3O 3I 3E - 7E 7E 7E U The line 3O 3I 3E 7E - 7E 7E U The spike 3O 3I 3E 7E 7E - 7E 3E Kitting 3O 3I 3E 7E 7E 7E - 3E Loading 3O 3O 3I U U 3E 3E -

57

presents the same type of information, but with respect to the plastic bins used in both warehouses. The plastic bins contain items that are essential smaller com-pared to other items stored in pallets. For this reason, the plastic bins do not repre-sent the items requiring the largest storage area, but it is however interesting to present this information to gain a holistic view of the utilisation rate of the whole storage area available. The tents are used for temporary storage. In one of the tents, tent 1, a large amount of items and pallets are stored in front of other pallets with material, which hampers the capability to handle the material in a desirable way. The total amount of positions, items stored on floor space and the utilisation are presented in table 21. To gain an understanding of the utilisation rate for the whole site, table 22 presents the total amount of available pallet positions, the total amount of occupied pallet positions and the utilisation rate. The pallets stored on the floor in one of the tents are disregarded in that context. To summarise this section, it is interesting to highlight that the utilisation rate for the whole site cor-responds to 81%.

Table 19 - Presents the total amount of pallet positions in both warehouses and the utilisation

Warehouse Total amount of pallet posi-tions

Occupied pallet positions Utilisation

A 3,850 3,031 78.73% A: Storage at KIT and SPIKE 212 148 69.81%

A: Frame storage 99 93 93.94%

SUM A 4,161 3,272 78.63%

B 2,484 2,086 83.98%

SUM 6,645 5,358 80.63%

Table 20 - Presents the utilisation of plastic bins in the two warehouses

Warehouse Total amount of plastic bins Occupied plastic bins Utilisation

A 270 220 81.48%

B 125 120 96.00%

SUM 395 340 86.08%

58

Table 21 - Presents the utilisation in the three tents

Tent Total amount of pallet positions

Occupied pallet positions

Floor stor-age Utilisation

1 162 162 84 152% 2 1,167 943 - 81% 3 377 293 - 78% SUM 1,706 1,398 84 87%

Table 22 - Summarises the utilisation for storage in the two warehouses and in the three tents

Warehouse Total amount of pallet posi-

tions Occupied pallet positions Utilisation

A 4,161 3,272 79% B 2,484 2,086 84% Tent 1 162 162 100% Tent 2 1,167 943 81% Tent 3 377 293 78% SUM 8,351 6,756 81%

4.4 Benchmarking 4.4.1 In-time Delivery Accuracy The average in-time delivery accuracy during one year for warehouse A is 97.3 %. This means in 97.3 % of the cases products are delivered in-time. Warehouse B has an average delivery accuracy of 94.3 %. The information for warehouse B is based on a data file received from one of their customers. The file contains data of delayed deliveries from warehouse B during the period of January 2013 to March 2014. In table 23, information of why the deliveries were delayed is to be found. A stated goal for warehouse B is to deliver 97 % in time and it applies to all custom-ers.

59

Table 23 - Reasons for delays for warehouse B

Reason for delay No. of late order lines Machine problem 79 Planning 73 Tool problem 39 Late delivery from supplier 34 No information 29 Defects 22 Short of personnel 21 Short of material 19 To early delivered from manufacturing departments 18 Correct delivery date according to information system 16 Short of capacity 12 Unknown 10 Registration problems 6 On-hand accuracy 4 Quality problem with supplier 3 Reprocessing 2 SUM 387

4.4.2 Picking Accuracy Data for warehouse A are collected from January 2013 to February 2014. In aver-age the picking accuracy is 99.91 % during this period. Incorrect deliveries of gaskets from suppliers affect the picking accuracy in warehouse B. Data concern-ing picking accuracy for warehouse B are presented in table 24. Table 24 - Picking accuracy warehouse B

No. of orders picked 2013-11 – 2014-03 8390 No. of picking errors for sheets 72 No. of picking errors gaskets 75 No. of picking errors others 8 No. of picking errors total 155 Picking accuracy 98 %

4.4.3 Receiving Accuracy and Supplier Quality Receiving accuracy is 97 % for both warehouses; this means that the suppliers deliver 97 % in time. The information is gained form an information system at Alfa Laval. A list of products received with poor quality is summarised. The data set is from January 2013 to March 2014. In total 1,633 products were delivered with poor quality during this time. The distribution between suppliers could be seen in figure 33. In the figure suppliers representing the largest share of errors are visualised. Of the products delivered with poor quality, 38 % were also delivered at the wrong time. Order lines could be delivered both earlier and later than the required date.

60

4.4.4 Picking Productivity In warehouse A personnel are picking about 24,500 order lines per month. Weight and dimensions of the product affect the time for picking one order line. One workday is seven hours. Four to five persons are working with the picking opera-tion every day, two persons with external customers and one to two persons with kitting. Three persons are working with inspection indoors every day and one per-son is working with inspection outdoors. Productivity is calculated as the lines picked per person hour and includes all warehouse personnel, only person hours dedicated to value adding activities are excluded. Productivity for warehouse A:

𝐿𝑖𝑛𝑒𝑠  𝑝𝑖𝑐𝑘𝑒𝑑/𝑚𝑜𝑛𝑡ℎ4

7  ×𝑃𝑒𝑟𝑠𝑜𝑛𝑠  𝑤𝑜𝑟𝑘𝑖𝑛𝑔  𝑤𝑖𝑡ℎ  𝑝𝑖𝑐𝑘𝑖𝑛𝑔/𝑑𝑎𝑦×5=

24,5004

7× 4.5 + 2 + 1.5 + 3 + 1 ×5

≈ 15  𝑙𝑖𝑛𝑒𝑠  𝑝𝑖𝑐𝑘𝑒𝑑/ℎ𝑜𝑢𝑟

In warehouse B 2,000 lines are picked per week. Approximately 18 % of the order lines picked are to external customers and 82 % of the lines go to internal custom-ers. 50 % of the orders are external and 50 % are internal, this means that internal orders include more lines. They are twelve persons working in warehouse B seven hours a day. Four persons are working with the picking operation, two with in-spection and one person is working full time with hole punching. From this data productivity are calculated for warehouse B:

Figure 33 - Suppliers delivering products with poor quality

61

𝐿𝑖𝑛𝑒𝑠  𝑝𝑖𝑐𝑘𝑒𝑑  /𝑤𝑒𝑒𝑘7×𝑃𝑒𝑟𝑠𝑜𝑛𝑠  𝑤𝑜𝑟𝑘𝑖𝑛𝑔  𝑤𝑖𝑡ℎ  𝑝𝑖𝑐𝑘𝑖𝑛𝑔/𝑑𝑎𝑦  ×5 − ℎ𝑜𝑢𝑟𝑠  𝑠𝑝𝑒𝑛𝑡  𝑜𝑛  ℎ𝑜𝑙𝑒  𝑝𝑢𝑛𝑐ℎ𝑖𝑛𝑔

=2,000

7×12×5 − 7×5≈ 5  𝑙𝑖𝑛𝑒𝑠  𝑝𝑖𝑐𝑘𝑒𝑑/ℎ𝑜𝑢𝑟

4.4.5 Warehouse Cost Warehouse cost is another interesting matter for this research. The cost for ware-house A and tent 3 during one year is 29,029,000 SEK and the cost for warehouse B and tent 1 and 2 is 32,335,000 SEK. The cost includes rent, salaries and other supplies. From this information it is possible to calculate the cost for each pallet position per day, see equations below. The cost is calculated for both warehouses together and for each of the warehouses. The cost per pallet position is higher in warehouse B than in warehouse A, the cost differs by about 6 SEK. The cost dif-ference probably depends on what activities those are included in the warehouses and also of the equipment used. In warehouse B no barcode system is used which most likely makes the material handling more expensive and inefficient than in warehouse A. Another reason could be that different activities are included in the two warehouses. For example warehouse B has personnel working with material supply and handling of raw material that warehouse A does not have. These activi-ties add costs to warehouse B.

𝑇𝑜𝑡𝑎𝑙  𝑐𝑜𝑠𝑡  𝑓𝑜𝑟  𝑡ℎ𝑒  𝑤𝑎𝑟𝑒ℎ𝑜𝑢𝑠𝑒𝑠  𝑓𝑜𝑟  𝑜𝑛𝑒  𝑦𝑒𝑎𝑟𝑁𝑢𝑚𝑏𝑒𝑟  𝑜𝑓  𝑝𝑎𝑙𝑙𝑒𝑡  𝑝𝑜𝑠𝑖𝑡𝑖𝑜𝑛𝑠  𝑎𝑣𝑎𝑖𝑙𝑎𝑏𝑙𝑒      

365=

61,364,0008,351365

≈ 20  𝑆𝐸𝐾  𝑝𝑒𝑟  𝑝𝑎𝑙𝑙𝑒𝑡  𝑝𝑜𝑠𝑖𝑡𝑖𝑜𝑛  𝑎𝑛𝑑  𝑑𝑎𝑦

Cost per pallet position for warehouse A:

𝐶𝑜𝑠𝑡𝑠  𝑤𝑎𝑟𝑒ℎ𝑜𝑢𝑠𝑒  𝐴𝑃𝑎𝑙𝑙𝑒𝑡  𝑝𝑜𝑠𝑖𝑡𝑖𝑜𝑛𝑠  𝑖𝑛  𝑤𝑎𝑟𝑒ℎ𝑜𝑢𝑠𝑒  𝐴  𝑎𝑛𝑑  𝑡𝑒𝑛𝑡  3

365=

29,029,0004,538365

≈ 17.5  𝑆𝐸𝐾

Cost per pallet position for warehouse B:

𝐶𝑜𝑠𝑡𝑠  𝑤𝑎𝑟𝑒ℎ𝑜𝑢𝑠𝑒  𝐵𝑃𝑎𝑙𝑙𝑒𝑡  𝑝𝑜𝑠𝑖𝑡𝑖𝑜𝑛𝑠  𝑖𝑛  𝑤𝑎𝑟𝑒ℎ𝑜𝑢𝑠𝑒  𝐵, 𝑡𝑒𝑛𝑡  1  𝑎𝑛𝑑  2

365=

32,335,0003,813365

≈ 23.2  𝑆𝐸𝐾

62

5 Identifying Areas of Improvement

This chapter presents potential areas of improvement that were identified by using the different concepts and from the identification of the current state. Identified gaps are first presented and subsequently improvements that could be made. The structure is: VSM, Activity Profiling and Benchmarking.

5.1 The Warehouse Performance Gap Analysis The warehouse performance gap analysis identifies and visualises the gap between a warehouse performances and world-class standards and thereby it also shows what areas that could be improved in the warehouses. The performance gap analy-sis is based on data from the previous chapter, identifying the current state.

For warehouse A and B the result is presented in figure 34. For most of the activi-ties both warehouses are performing to the same level. At receiving they are un-loading, staging and checking arrived material and items, stage 1. For put-away the prevailing situation is first-come-first-served corresponding to stage 1 and at the reserve storage conventional racking bins are used, stage 2. Both warehouses are applying pick-to-single-order, stage 1, and by slotting the planning is random, stage 1. The replenishment activity occurs when needed and when the pallet at the

Figure 34 - Warehouse Performance Gap analysis for warehouse A and B

63

default position is totally empty for both warehouses, stage 1, and at shipping or-ders are checked, staged and loaded, stage 1. For work measurement the minimum distance to world-class appears; both warehouses are performing at stage 3 that implies that standards are used for evaluation. For communication warehouse B is using paper, stage 1, and warehouse A are using RF terminals, stage 3.

5.2 VSM The VSM visualised where material and items were waiting, either to get stored in the main storage or at the customers’ areas. Table 25 summarises, for both ware-houses, how many orders that were waiting. Waiting orders include products that were overproduced, picked and transported to the customer but not used by the customer right away. The areas occupied for waiting are also summarised in the table with the purpose to highlight how much space that is occupied. A stands for warehouse A and B for warehouse B.

Table 25 - Waiting and overproduction identified for both warehouses

In a future state the areas where material and items are waiting should be eliminat-ed and material should after arrival to the site be put directly into the warehouse. Orders from the warehouses should only be delivered when the customer request it. Continuous flows and pull-systems should be implemented; the material flows should be continuous from arrival to departure. The continuous flows are achieved by producing what is needed in the next stage, when it is requested. By imple-menting these changes, no products have to wait and unnecessary starts and stops are avoided. This is an important finding for Alfa Laval and should definitely be considered when presenting a suggestion for a future state and recommendations. In warehouse A the remaining steps should be inspection, put-away, picking and packing. These steps are considered to be the only steps necessary in a warehouse. The same concerns warehouse B. The two figures, 35 and 36, show what steps that should remain in the warehouses according to the VSM. The VSM shows what areas to focus on. The crosses in the two figures, 35 and 36, show what steps that should be removed in each of the two warehouses. If the crossed steps are not possible to remove they should be replaced with a continuous flow or a pull-system. The largest storage areas where products are waiting the longest time

Waste Snapshot (orders) Area (m2) Waiting (A) 103 250 Waiting (B) 20 250 SUM 123 500

64

should be removed first. With these improvements about 500 m2 could be released in or outside both warehouses.

From the mapping distances for each of the material flow were documented. In the same way as for waiting and overproduction, transportations of material and items between different locations and activities at the site are considered to be wasteful. This information gained from the mapping is an important background when con-sidering how activities should be placed in the warehouse and also where the warehouse should be located at the site in order to enable short transportation dis-tances. The transportation distances cannot be eliminated, but should however be an important aspect when evaluating a new suggestion with the purpose to reduce them. By merging the two warehouses to one building, transportations between the

Figure 36 - Steps that should be removed in warehouse A according to the VSM

Figure 35 - Steps that should be removed in warehouse B according to the VSM

65

two warehouses will be eliminated. By placing the unloading dock closer to the warehouse, transportations will be reduced. The material flows within the site could also be improved in order to reduce distances.

Tables are missing at inspection and therefore motion was identified while map-ping. Inspection is not only causing motion, it contributes also to over processing that is considered as a waste. Three to four persons in warehouse A and two to three persons in warehouse B are working with inspection seven hours a day. In-spection is the root of two of the wastes. It should be evaluated how the inspection could be improved in order to reduce the over processing and also motion in both warehouses. This is another area of improvement identified from the VSM. An-other important finding from the VSM was how many personnel that are involved in the activities along the material flows and how much equipment that are used to perform the activities. This information gives a good foundation for where poten-tial savings are to be found. Inspection is a problem area that was identified by the VSM but also an opinion expressed by the personnel. Other opinions expressed by the personnel were that some suppliers are not meeting the requirements set by Alfa Laval and that late certificates are causing occupied space by pallets that can-not be released for picking. Most of the opinions mentioned by the warehouse personnel for both warehouses are areas for improvement and should be taken into consideration when deciding on actions and long-term goals (VSM with ware-house A and VSM with warehouse B).

5.3 Activity Profiling The fact that some products have higher frequency than others is vital for the deci-sion on where to place the products in the pallet racks. High frequency products should preferably be stored in pallet positions near to the floor, level one and two, enabling effective handling operations. It would also be interesting to investigate how a forward picking area could be implemented at Alfa Laval for high frequen-cy products. In the two figures, 37 and 38, it is possible to see where high frequen-cy products should be stored in each warehouse. Picture 37 shows how high fre-quency products should be placed in warehouse A. The arrows mark the exits. Products arriving from exit 1 and frequently departing through exit 2 should pref-erably be stored at area C. Products arriving through exit 1 and departing through exit 3 should preferably be stored in area D. High frequency products that arrive through exit 1 and often depart through exit 1 should be placed in area E. High frequency products arriving through exit 4 and departing trough exit 3 should be placed in area B. Products should be stored in area A if they arrive through exit 4 and depart trough exit 1 and products arriving through 4 and departing through 2 could be stored in either area A or D. In warehouse B, figure 38, high frequency products arriving through exit 1or 2 and often departing through exit 3 should

66

preferably be stored in area B or C. High frequency products arriving through either exit 1 or 2 and departing through exit 1 or 2 should preferably be stored in area A.

It was identified that it would be beneficial for some of the products to be deliv-ered in more appropriate cases, containing an amount that corresponds to the quan-tity that the customer most of the times orders. It will not only increase the effi-

1 2

3

A

B

C

D

E

4

Figure 37 - How to store products warehouse A

2

1 3

A

B

C

Figure 38 - How to store products warehouse B

67

ciency when picking, it would also decrease picking errors for the warehouses and eliminate the human factor impacting on the picking accuracy.

From the customer order profiling, it was identified that most of the orders ordered by the customers are single line orders for warehouse B. Based on this information it can be concluded that there is no reason for investigating which products that most of the times are ordered together, since most of the products are not ordered together with any other product at all for warehouse B. However, it could be of interest for Alfa Laval to investigate how single line orders could be batched to-gether in order to reduce the transportations in the warehouses when several orders can be picked at the same time. In warehouse A there are almost as many single line orders as multiple line orders. Here it would be interesting to investigate both how orders could be batched together and if products tend to be requested together and if that would impact how products should be stored in the warehouse.

5.4 Benchmarking One area that was identified already on an early stage as a potential area for im-provement was the charge clean pallet. If pallets could be utilised in a more bene-ficial way how would that impact the amount of occupied pallets and how many pallet positions could be released? A file was received from warehouse A contain-ing information about the number of pallet positions, the type of product stored in the pallet and the quantity stored in every pallet. With this file it was estimated how many storage locations that could be released if the products were stored more efficiently. An example is shown in table 26.

Table 26 - An example of a product that is stored in six different locations

Pallet position Quantities stored Receiving date 1 5 2014-02-11 2 2 2013-12-17 3 8 2013-12-17 4 5 2014-02-11 5 5 2014-02-11 6 6 2014-02-11

The same type of products is stored in six different pallets as a result of charge clean pallets. It is assumed that all pallets have the same size and that each pallet can store the same amount of pieces. In this case each pallet could store eight pieces corresponding to the greatest amount identified. Based on this assumption, the number of pallets that actually are required to store this product could be esti-mated:

68

5+ 2+ 8+ 5+ 5+ 68 = 3.875

The result shows that four pallets are needed instead of the six that are currently used; it is possible to release two pallet positions. This was done for all products and the result shows that at least 142 pallet positions could be released in ware-house A. In the same way storage locations at the yard were analysed and the con-clusion was that 31 pallet positions could be released at the yard. In warehouse A there are 4,161 pallet positions in total. This means at least 3.4 % of the storage locations could be released.

1424,161 = 3.4  %

A pallet position costs 20 SEK per day. This means that 142 pallets correspond to 1,036,600 SEK in a year in terms of money. Products that are currently stored at the yard could be taken inside. Those 142 products that are currently stored at the yard and that could be taken inside would not be damaged by rust and the picking operation would be more efficient as a result of reduced distances. No conclusions could be made about the products stored in only two pallets. There are 466 prod-ucts stored in two pallets that mean there could be 233 more pallets to release if it is investigated more into depth. This is a potential area of improvement for Alfa Laval to consider.

From the current state some reasons for delay from warehouse B to one of its larg-est customer were identified. A number of these reasons are factors that cannot be impacted by the warehouse organisation and shall therefore not be the focus. In-stead areas of improvement shall be identified based on the things that warehouse management actually can impact. In table 27 these factors are marked with a zero since it is assumed that these reasons for delay shall be eliminated. At the moment the warehouse delivers 94.3 % in time and their goal is to deliver 97 % in time. What happens if we eliminate all the defects warehouse management can impact? During the period that data in the file were recorded, 6,790 lines were sent to the customer from warehouse B. If all the reasons for delays that warehouse manage-ment and personnel can impact would be set to zero the warehouse would deliver 96.4 % in time, see equation below. So even if warehouse management perform their best they are not able to reach their goal.

6,790− 2446,790 = 96.4  %

69

Table 27 - Reasons for delays to one of warehouse B’s customer

Reason for delay No. of late order lines Machine problem 79 Planning 73 (0) Tool problem 39 Late delivery from supplier 34 No information 29 Defects 22 Short of personnel 21 (0) Short of material 19 (0) To early delivered from manufacturing departments 18 (0) Correct delivery date according to internal information systems 16 Short of capacity 12 (0) Unknown 10 Registration problems 6 On hand accuracy 4 Quality problem supplier 3 Reprocessing 2 SUM 387 (244)

5.4.1 World-Class Measures There are a number of measures important for a warehouse. A handful of these are presented in table 28 (Innovation and Best Practice for Business Success). The best in class definition is presented, so is the median and finally the actuality for the warehouses in this case study. The aim is to position the warehouses against what is considered as world-class in order to improve the existing measures in the warehouses. The warehouses at Alfa Laval should constantly strive to improve performance, this even though it may not be realistic to reach the world-class per-formance considering their warehouses. Warehouses are although not a core busi-ness at Alfa Laval.

Table 28 - The two warehouses are positioned against world-class

Category Best in class Median A B On-time shipments ≥ 99.8 % 98.70% 97.3 % 94.3 % Picking accuracy ≥ 99.9 % 99.50% 99.1 % 98 % Lines picked and shipped per hour ≥ 74.2 28 15 5 Receiving accuracy ≥ 99.5 % 98.5 % 97 % 97 % Safety (accidents per year) 1 5 0 0

70

6 Recommendations

This chapter presents the recommendations that have been identified for Alfa La-val in order to improve the current state. It is structured according to how each concept has contributed. As a final part of this chapter a suggestion for the future state is presented and how each of the concepts have contributed to this sugges-tion.

6.1 VSM Based on identified areas of improvement it is recommended for Alfa Laval to consider a solution where arriving material and items are directly put in the main storage without any time spent on waiting. All orders that are overproduced, or-ders delivered to the customer before they are requested, should be eliminated; orders are picked and delivered to the customer when the customer needs them. To enable the elimination of overproduction a continuous flow from the warehouse to the customer at the site and the external customers should be applied.

At the site material flows are covering large distances and material and items are transported unnecessary long distances. One major cause for this problem is the localisation of both warehouses. At the moment there are two warehouses and material and items are transported between these two units. If these two units could be merged this could improve the physical material flows at the site and transportations between the warehouses could be eliminated. To merge the two warehouses to one unit and place it close to the arriving gate is a recommendation to Alfa Laval. The existing waiting areas could be eliminated if Alfa Laval located the warehouse near to the gate for arrival and departure.

Except the importance to locate the warehouse near to the arrival and departure gate, it should also be close to the two other buildings at the site with production, assembling and processing. This in order to avoid excessive transportations and handling of material when material goes from production to the warehouse and subsequently to the assembling unit. If the warehouse could be located in connec-tion to the two other buildings, preferably between these two, it would enable a continuous flow where starts and stops are eliminated.

Another advantage of merging the two warehouses is that similar activities in both warehouses could be merged to one activity e.g. handling of external orders. This

71

could release personnel involved in one specific activity and equipment needed could also be reduced. It would improve the planning unit for these activities.

Inspection is an activity identified as a station causing problems in the warehouse organisation. A recommendation would be to investigate how this activity could be eliminated or how the importance of the activity could be reduced. To decrease the time spent on inspection Alfa Laval should work with their internal inspection processes securing that drawings are correct and products should only be inspected when there is a need for it.

6.2 Activity Profiling An improved warehouse layout for Alfa Laval would be to place the high frequen-cy products near the floor and divide the warehouse into zones with respect to the type of product; similar products should be placed in one zone. The bird’s eye identifies where high frequency and low frequency products are stored in the cur-rent warehouses and from this information improvements for a future state can be identified. Another recommendation to Alfa Laval is to negotiate with the suppli-ers about products that can be delivered in cases with quantities better correspond-ing to how the customer orders from the warehouses. This could reduce picking errors and simplify the manual handling of picking for warehouse personnel.

The activity relationship profile identified the dependency between activities in both warehouses. The most important dependency identified was the correlation between activities sequencing each other; a preceding activity impacts the se-quencing activities. One identified area of improvement was that activities in the end and the beginning of the flow should be localised near to a gate enabling shorter transportation distances. As said in the previous section, activities can be merged. For warehouse A, the spike and kitting may be merged, since the activi-ties are almost identical. This could release space in the warehouses, streamline the physical material flows and prevent that identical activities are done twice. Since a recommendation is to merge the two warehouses, the kitting, the spike and the external activity for warehouse B can be organised into one activity.

Since there are more frequently picked products, it should be investigated where these products are located in the warehouse. Bird’s eye will be helpful for this objective to identify the most convenient pallet positions in the warehouses and give a good foundation for decisions.

72

6.3 Benchmarking The best in class warehouse utilisation identified is greater than 90 %. It even exist warehouses with an utilisation of 100 %, but they are not as efficient and produc-tive as desired. Frazelle (2002a) recommends a maximum utilisation of 86 %. As can be seen from the identification of the utilisation in the current state, the utilisa-tion rate at Alfa Laval is at an equivalent level. For this reason it would be recom-mended for Alfa Laval to keep the utilisation rate at this level and not make any major changes in this area. Another recommendation is to implement a tool for measuring the utilisation on a regular basis in order to investigate if the utilisation changes over time. The utilisation rate is an important measure for Alfa Laval to keep track on.

Warehouse B has an in-time delivery goal of 97 %. In the section identifying areas of improvement it was calculated that this goal could not be reached even though warehouse management eliminated all reasons for delay that they can impact. Is this goal then an achievable goal for the warehouse and something to strive for? A recommendation to Alfa Laval would be to eliminate all late deliveries that are due to factors that warehouse management actually can impact. To achieve this goal delayed deliveries should be recorded as well as the reason for lateness. Ware-house management should try to improve planning and at the same time it should be warehouse management’s responsibility to inform other groups about defects affecting the warehouse performance.

For the current state it was calculated how much every pallet position costs per day. The aim was to calculate how much every pallet position costs every day at the site in Lund and the result was 20 SEK. A difference of 6 SEK between the two warehouses was also identified This information is partly important in order to get an idea of how much it costs to store material and products in the warehouses and the tents, and also when evaluating different alternatives to reduce costs. It is important to be aware of the actual costs when evaluating different options to re-duce costs that may involve large investments. It also gives an indication on how costly it is to store material for a longer period and if it is worthwhile to try to release pallet positions in the warehouses. Pallet positions could be released by storing products with different charge numbers in the same pallet, which would require pallets that could be divided into zones and also a reconstruction of the information system. It is recommended to take this finding into consideration when evaluating potential improvements.

Alfa Laval should work to increase the warehouse productivity, but the responsi-bility to achieve this should be taken by management at a higher level. For ware-house personnel, team leaders and unit managers the goals have to be broken down to measurable indicators. One goal for warehouse B could be to not have any late deliveries that are due to lack in planning.

73

6.4 The Combination of the Concepts Alfa Laval should work with their supplier processes. Some suppliers deliver only on certain days in the week, which makes the workload and inventory levels une-ven. Receiving should therefore be more continuous and Alfa Laval should be more demanding concerning the receiving accuracy. That delivery is done only on specific days from some suppliers was discovered during the VSM and Bench-marking discovered the receiving accuracy. The Activity Profiling presented in-formation on how the case quantities can be reorganised by the suppliers.

For warehouse B one area is causing a lot of the problems, namely the planning. By adopting the VSM, possibilities for pull-system and continuous flow can be found that would simplify planning. These systems will make planning easier and will also straighten the flows in the warehouses. A prevailing problem area for both warehouses is long transportations and many stops along the way. This could be solved partially by using the VSM and the relationship activity profile, to know where stops could be avoided and what activities that should be placed closer in order to reduce distances. Alfa Laval should start by doing the changes proposed in this thesis and then map the current state again to see what improvements that should be done next.

Warehouse B should use the same barcode system as warehouse A in order to reduce picking errors. The system would decrease the time spent on searching for products and picking errors would be reduced. This is a fact recognised by com-paring both warehouses against each other; warehouse A has a better delivery accuracy compared to warehouse B and the reason for this is derived from the application of a barcode system in warehouse A. The information systems at Alfa Laval should be up to date so that the utilisation could be kept at a constant level to increase productivity. Activity Profiling and Benchmarking were essential to understand the importance of the utilisation rate in a warehouse. Alfa Laval should also focus on dividing their pallets into sections in order to store more products in their warehouse, identified by the Activity Profiling.

6.5 A Suggestion for the Future State Based on the framework, criteria have been identified to create an ideal future state. All recommendations presented in this chapter are considered for the future state and are summarised for respectively concept. From VSM some criteria were identified as more important than others for a final suggestion of the future state. Firstly, to reduce the amount of duplicate work, enabling one planning unit and improve the material flows, the two warehouses and the supplementary tents should be merged to one warehouse, see figure 39. The warehouse should be

74

placed near the arrival gate so arriving material and items can be unloaded on a dock at the warehouse. Completed orders to external customers should also be loaded on the dock at the warehouse on trucks going out from the site. All starts and stops, material and items waiting and overproduced, should be eliminated. There should be a continuous flow from the moment material and items arrive to the site until they are delivered to the customer.

In the warehouse duplicate activities can be merged together and as a result ware-house personnel can be released from e.g. the inspection. The amount of people working with one activity can be reduced. As can be seen in figure 39, the ware-house is located between the both existing buildings as suggested. This will enable improved material flows at the site; processed material and items from the produc-tion unit could go directly into the warehouse and from picking orders are directly moved to the assembling and manufacturing unit. Transportations back and forth at the yard between the warehouses are eliminated and that is an important aspect for the future state.

From the Activity Profiling some findings are important for the proposed future state. The warehouse should be divided into zones with respect to the type of product enabling a warehouse within the warehouse, see figure 40. High frequency products should be placed near the floor in order to enable fast pick, this can also be evaluated based on the findings from the bird’s eye. There can be fast-pick areas within the zone. Inspection should be placed near the arrival gate reducing transportation distances for arriving material. Kitting, the spike and external orders in warehouse B, should be merged to one activity and be placed near the departure gate from the warehouse. Hole punching should be placed near the zone storage for plates. Gaskets and plates are localised in the warehouse near the gate to the

Figure 39 - A future state for the warehouse at Alfa Laval

Workshop

Warehouse Presses

Raw mate-rial

Free space

Frames manu-facturing

Departure

Manufacturing of small and medium

sized PHEs Manufacturing of large PHEs

75

production unit, where plates are processed. The line is localised near the gate where assembling with TAKT is located. Products mainly used for internal orders are located near the gate to the assembling unit and material and items most of the time ordered by external are located near to the activity handling external orders, see figure 40.

Benchmarking has identified some decisive information for a final suggestion how to reorganise the warehouses and the tents. The utilisation level in both ware-houses, and the tents except the one with an utilisation above 100 %, is at a desira-ble level corresponding to world-class measure. This implies that the new ware-house calls for the same amount of pallet positions used at the moment and there-fore the corresponding space used at the moment. Alfa Laval should work more with their supplier relations and continue to work with the receiving accuracy. Focus should be on the suppliers causing most of the problems. This would also help to reduce the importance of the inspection and focus can be on what is really causing the problems.

In a new warehouse the barcode system from warehouse A should be implemented in order to keep the picking accuracy at the same level as for warehouse A. Except this benefit it would also reduce the problem with searching for material and items since the barcode system record where material and items are stored. Goals such as in-time delivery to the customer should be set so they are obtainable for the

Gaskets and Plates The line

External Internal

Departure and Pack-ing

Arrival and Inspec-tion

Figure 40 - Warehouse layout inside the warehouse

76

organisation. Goals should also be broken down so that warehouse management can adapt them to their organisation and every day work. Measures for productivi-ty should be implemented and measured on a regular basis.

6.6 The Gap Analysis for a Future State The warehouse performance gap analysis has been used in this report to visualise the gap between the current state and what is defined as world-class. It can howev-er also be used in order to illustrate where the warehouse will be positioned after reengineering projects. In order to emphasise the benefits that can be gained with this suggested warehouse for the future, figure 41 presents the gaps between a future warehouse state and world-class. At receiving the arriving material and items would be put-away immediately to primary, stage 3, and put-away would be batched by zone, stage 2. Reserve storage would still be conventional racking bins, stage 2, and picking would be zone picking with downstream sorting, stage 4. Slotting would be at stage 3 implying popularity and cube based. Replenishment is still as needed (pick face complete) at the lowest level, stage 1. Shipping would be stage and load, stage 2, and work measurement would still be at stage 3. Commu-nications would be at stage 3 that implies RF terminals.

Figure 41 - Warehouse performance gap analysis of the future state

77

7 Analysis of Framework

In the analysis chapter the framework is analysed in order to answer the research questions. The analysis focuses on whether the framework including the three concepts VSM, Activity Profiling and Benchmarking, is useful to evaluate a ware-house and to identify the current state.

7.1 VSM In the frame of reference the product family matrix was explained as a vital part of the VSM. This step was however not performed during the study. The product family matrix was not useful at these circumstances and to include all the products stored in the warehouse in a matrix would require an extensive investigation. In the warehouses it was found to be more beneficial to identify the largest customer where material and items are more or less handled in the same way. The largest customer was identified by personnel working in the warehouse and the flow was followed counter current by asking about the preceding step. If there were two or more alternative preceding steps for one activity, the largest material flow was followed. This is also the recommended approach when applying this framework.

One aspect that can complicate the VSM is that it requires the involvement from personnel. Without personnel with good understanding and knowledge of the warehouse the VSM is not feasible and the result will not be representative. Dur-ing this study it was important to communicate the importance of why the VSM should be performed and to involve the right persons in the project.

One of the great benefits with the VSM is that data collected are primary data. Most data are collected when walking at the production site. The VSM can identi-fy for how long time material and items are waiting and how many orders that are picked before the customer actually requests it, overproduction. This is an im-portant finding identified by the VSM and this is also a reason for applying the VSM when investigating a warehouse. By identifying overproduction in the ware-house possibilities for pull-systems and continuous flows are recognised. With the VSM it is possible to identify wasteful steps in the warehouse, those steps where material is only waiting and therefore should be removed. The risk by removing the waiting areas closest to the manufacturing unit could be delayed deliveries of material and components to this unit. If that would become a reality it can in the long-term perspective cause stops in the production. In order to avoid this, the time an order is waiting directly before manufacturing should slowly be decreased.

78

Pull-systems and continuous flows could be used instead of removing the entire waiting area. There are no risks by e.g. removing the waiting area in the gateway and other locations that are not directly linked to the manufacturing unit. The VSM identifies how many personnel that are involved in different activities along the material flow and how much equipment that is needed as well. Thereby, it is identified where it is possible to reduce staff and equipment or where new equip-ment should be implemented. The VSM gives a good foundation when evaluating how a warehouse should be localised in order to enable the most beneficial trans-portation distances at the site.

One aspect that was not performed for the VSM was to visualise the timeline for a material flow. It was not feasible for a warehouse, since material and items are stored in the main storage for varying time. It may involve just a couple of hours or days to months that material and items are stored. If the timeline should be vis-ualised the material flows have to be divided into significantly smaller groups with respect to how long material and items are stored in the main storage.

In frame of reference wastes were defined and translated to warehousing. During the empirical study, the chapters about identifying the current state and areas of improvement, it was evaluated how well this translation correspond to the reality. It was discovered that most of the wastes could be found in a warehouse in ac-cordance to the translations. Overproduction was all orders picked and shipped before the customer actually needs it and material and items were found waiting extensive time before they reach and after they leave the main storage. As has been discussed there were plenty of transportations involving long distances. Ma-terial and items were over processed at inspection and inspection was also causing motion since it involves excessive movements for the personnel. Defects could also be identified; orders were delivered and received with wrong quantities, poor quality and at the wrong time. With the wrong time means both material and items delivered too late and too early than requested. Since wastes could be identified in accordance to the translation of waste for a warehouse, the translation can be con-sidered as valid. This is an important finding and is significant when applying this framework. Wastes found during the VSM, Activity Profiling and Benchmarking are as the definitions in table 4.

7.2 Activity Profiling The Activity Profiling contains a number of profiles and some of them were not applied and therefore not evaluated during this study. The inventory profile would actually be of interest since it provides the company with information about what is stored and for how long it is stored in the warehouse. This would give an exten-sive background in order to evaluate if the right products are stored and may result

79

in released pallet positions. This was however not performed since this profile is considered as rather time consuming and also because of the fact that Alfa Laval is already working with a complex ABC system. Calendar-clock profile is another profile included in the Activity Profiling that identifies seasonality in demand for products stored in a warehouse. This profile was neglected since this study is re-viewing a company with negligible seasonality for their product demand. It would nevertheless be an interesting area for further research. The investment profile has also not been applied, but here is definitely potential to investigate how this profile could contribute and especially how it would complement the analysis of how to reach a future state. The investment profile indicates costs and operating parame-ters necessary for the decision-making about investments. In this case it would be interesting to investigate how this would contribute to a better-grounded decision making in order to reach a future state. It would also be interesting to do a pur-chase order profile to see if the inventory levels could be held in a more constant level; even though, Alfa Laval already has a system that automatically tells when a product should be purchased. The activity relationship profile was in this case found to be useless since the VSM gives a better understanding of the relations between different activities in the warehouse.

One problem with the Activity Profiling is that the data analysed are secondary data collected from the internal information systems in the warehouse. The quality of the data, the data analysis and the outcome of those analyses are all dependent on the quality of the information systems. If the data collected from these systems are not complete the outcome of the analyses will not be complete either.

The Activity Profiling focuses mostly on put-away and picking operations in the warehouse. Those operations are important for the warehouse productivity and therefore the Activity Profiling is seen as an important part of the framework.

7.3 Benchmarking Some important findings were found by using Benchmarking. The utilisation rate for both warehouses and the supplementary tents were documented and compared and by using Benchmarking the utilisation rate could be compared to world-class utilisation. Quality measures, productivity and the cost for each pallet position were calculated. Some of the quality measures and the picking productivity were compared against world-class. The two warehouses were compared against each other concerning the costs for each pallet position and picking accuracy. All these aspects are important when evaluating warehouse performance. However, there were some difficulties and limitations with Benchmarking. For instance the world-class measures may not be applicable since there is no information about where the world-class numbers come from, how they were recorded or what kind of

80

products the warehouses handle. One example was the line picked per hour. A number of 74.2 lines picked per hour may be derived from a completely different branch and is not comparable with this case. An alternative to benchmarking against world class is internal benchmarking or benchmarking against companies within the same industry and warehouses handling similar products.

The advantage with the warehouse performance gap analysis is its ability to visu-alise the gaps between the current state and what is considered as an ideal state. Since it is the current state that is drawn in the warehouse performance gap analy-sis, information about this state is required in order to perform this part of the study. The information is collected during the VSM and the Activity Profiling. These two information sources provide enough information in order to identify the gaps to world-class for both warehouses. One problem identified with the ware-house performance gap analysis in this case was that some of the stages at higher levels were actually fulfilled by both or one of the warehouses, but since they did not fulfil a requirement at a lower level the warehouses were rated at an even low-er level. An example is warehouse B that fulfils stage 5 at work measurement, but since it does not fulfil the requirements for stage 4 they will instead end up at stage 3. The warehouse performance gap analysis is considered to be an important part of the framework, it is used to visualise the possibilities to improve the warehouse operations and it could also identify goals for a reengineering project as can be seen in the chapter about recommendations.

7.4 The Framework One problem found with using the framework is that it is time consuming. This framework would mainly be recommended to a company interested in investigat-ing the current state and wants to know where to put focus for making improve-ments. A company already aware of where to put focus may not find this frame-work as beneficial. The advantage with using the framework is that it gives a ho-listic picture of the warehouse and that the framework helps to identify areas for improvement. Figure 42 describes how the framework should be applied. The VSM and the Activity Profiling give proposals of how the currents warehouse state could be improved in order to reach a future warehouse state. In the future warehouse state wastes are reduced and warehouse operations are improved. Benchmarking is used to measure performance to see if the warehouse operations actually did improve and if the wastes were reduced by the proposals given by VSM and Activity Profiling. In frame of reference VSM, Activity Profiling and Benchmarking were presented. Two of the research questions stated in the intro-duction are whether the concepts complement and overlap each other. To be able to answer these questions in the conclusions the concepts have to be compared

81

against each other. In figure 42 it is discussed how the concepts overlaps and complements each other.

Figure 42 - Describes how the framework should be applied

Benchmarking can identify warehouse defects better than the VSM. Warehouse defects are e.g. wrong quantity picked, late delivery of order and wrong product picked. Those errors lead to rework and therefore they count as defects in a ware-house and can be identified by Benchmarking. This is because Benchmarking can compare the current state against other warehouses and world-class in order to set goals for the future state. It is also possible with Benchmarking to get ideas from other warehouses of how these defects could be reduced. Therefore Benchmarking than with the VSM better identifies defects.

Performance measures, derived from Benchmarking, can mainly complement Ac-tivity Profiling by providing suitable ways to present the information identified by the Activity Profiling. E.g. if products are placed according to the customer order profiling, it is possible to know if the products are actually placed in a more effi-cient way by measuring productivity. By using performance measures, improve-ment of performance can be measured and information can be gained whether the attempt to improve the warehouse has been successfully.

Basic information such as area of the warehouse and equipment used was collect-ed from the actuality, consequently from the VSM. Other basic information such as number of order lines picked was collected as a part of the Activity Profiling from an information system. Basic information can be gained from the VSM, but it can likewise be gathered from an information system, the Activity Profiling. It is

Current State

Warehouse operations

Wastes

Future State

Warehouse operations

Wastes

Measure and validate performance

VSM and Activity Profiling are applied to the current state and give proposals of how the cur-rent state could be improved.

Improved

Reduced

82

however recommended that all information that can be gained from the actuality should be collected as a part of the VSM instead of as a part of the Activity Profil-ing. Not only because it is convenient to gather this information when mapping, but also because it is a primary source instead of a secondary source of data. Basic information from the current state will also help to decide on how a future state should look like. The Activity Profiling will help to reduce transportations and inventory in a future state. The calendar-clock profile will facilitate scheduling e.g. how the seasonality of a product affect work force needed. The investment profile has not been used in this study but will probably give important information of where to invest in order to reach the future state.

83

8 Conclusions

In this chapter the research questions are answered based on the literature re-views and the analysis of the framework. Proposals for future research areas are also presented.

In this study an extensive investigation has been accomplished in order to evaluate if a framework, based on the three concepts VSM, Activity Profiling and Bench-marking, is suitable to present the current situation in a warehouse and to identify areas for improvement. The main purpose is to investigate how these individual concepts overlap and how they complement each other. This was examined through the Alfa Laval case. It is also answered what benefits that were found with using VSM in a warehouse, since the VSM is a concept not commonly used in a warehouse.

8.1 Answering the Research Questions RQ1: How do the three concepts complement each other and why should they be used together in order to evaluate a warehouse?

During this study it has been a discussion back and forth in that order each concept shall be performed. The order that has been applied in this study and that will be recommended is to start with the VSM, continue with Activity Profiling and final-ly complete the study with Benchmarking.

The initial mapping provides a snapshot of the current state and visualises the physical material flows. It identifies where material and items are waiting and where overproduction occurs. It identifies transportation distances and how many personnel that are involved in different activities and the equipment required. The area of the warehouse could also be identified during the VSM. In this study it is recommended to collect as much information as possible from the actuality during the VSM. Information that could not be collected from the actuality will be col-lected from information systems as a complement to the VSM. The next step is therefore to perform the Activity Profiling since it complements the findings from the VSM. It provides a background for the discussions and the decision-making concerning how to locate products in the warehouse. The VSM will not give any immense data set with information about how e.g. customers order from the ware-houses or if products tend to be requested together. The Activity Profiling does

84

this, which was an important aspect when it was decided to apply this concept for this study.

Both the VSM and Activity Profiling can identify the actuality, but neither of them can determine if the actuality corresponds to a desired situation, therefore measures for assessment are wanted and that motivates the Benchmarking. With Benchmarking it is possible to define goals for the future state. It is possible to identify existing gaps between what is referred to as world-class standard and the current situation in the warehouse. If there is nothing to compare against, it will be difficult to find objectives that motivate improvements for a future state. VSM and the Activity Profiling can be seen as crucial parts in order to obtain the warehouse performance gap analysis, which in turn is important in order to understand the gaps.

Figure 43 illustrates the order that each of the concepts should be performed when applying the framework and which waste that was identified by each concept. The VSM identifies all kind of wastes in a warehouse except inventory, motion and

Figure 43 - Wastes identified by each concept and the order that the framework should be performed

Activity Profiling Provides detailed

information about the activities involved in

the warehouse

Benchmarking Can identify the gap to world-class standards and define measurable

goals

VSM A snapshot of the

actuality and a map of the material flows

Over processing

Waiting

Transportation

Overproduction

Motion

Inventory

Transportation

Defects

WASTES

85

defects. When the VSM is done it includes taking notes of how much time that is spent on each activity, to count products that are waiting, measuring distances between the different activities in the warehouse, counting products that are picked but not used directly by the customers, named overproduction and also to take notes of equipment used. By using the Activity Profiling it is possible to evaluate inventory levels, this is done by using the inventory profile e.g. the ABC analysis. Using the Activity Profiling, by placing high frequency products in the most con-venient areas, could reduce transportations. With Benchmarking it is possible to define defects in a warehouse. Defects in a warehouse are picking errors, delivery errors or receiving errors and those are considered to be wastes. In this aspect Benchmarking is a more comprehensive way to identify defects in a warehouse compared to VSM. By a VSM it is not possible to know what is considered as a good number and therefore Benchmarking is preferred and recommended in this case.

Besides the identification of wastes, the framework evaluates the warehouse op-erations. In figure 44 it is illustrated which of the concepts that evaluate the differ-ent operations in the warehouse. Benchmarking evaluates all the operations since it defines measurable goals in order to reach world-class performance e.g. receiv-ing accuracy and picking accuracy. VSM evaluates all of the operations and how they relate when mapping. Activity Profiling takes many aspects of the warehouse

Receiving Put-away Storage Picking Shipping

Activity Profiling identifies areas of improvement. The Activity Profiling takes many aspects into account but has a focus on put-away, storage and picking. It identifies how material and items should be stored in a more beneficial way in order to simplify these activities.

Benchmarking can identify the gap to world-class for all of the activities and define measurable goals to strive for.

WAREHOUSE OPERATIONS

VSM identifies areas of improvement for all warehouse opera-tions as well as between the operations.

Figure 44 - Which of the concepts that evaluate the different warehouse operations

86

into account, but when applying this framework focus is mainly on put-away, stor-age and picking. It is recommended to collect basic information during the VSM if possible. Activity Profiling can identify how put-away, storage and picking could be organised based on circumstances such as how the customer orders and which products that are high frequency products.

The three concepts complement each other also in other ways. The current state could be compared with the future state in a gap analysis. The performance measures were important in order to get a general view of the current state and to understand what measurements those are important for a warehouse. With Bench-marking, performance measures, it is possible to perceive improvements of per-formance gained by the Activity Profiling and the VSM. Some of the basic infor-mation that is a part of the Activity Profiling was collected during the VSM such as area of warehouse and equipment needed. The investment profile tells how to invest in order to reach the future state.

RQ2: How do the three concepts overlap and were there any complications with using them together?

All three concepts can identify wastes; the VSM when mapping the current state, Activity Profiling when analysing secondary data and Benchmarking when com-paring the quality measures against world-class standards. There is only one waste overlapping, namely transportation that is identified by both VSM and Activity Profiling. By the VSM transportations are measured between the different opera-tions in the warehouse and with Activity Profiling transportations are reduced within two of the warehouse operations, namely put-away and picking. One area within VSM and Activity Profiling that does overlap is the identification of how activities depend on each other. This aspect is important for both concepts; for the VSM it is a natural result of the mapping and for the Activity Profiling it is accomplished with the activity relationship profile. The VSM was also used as a base when the activity relationship profile was completed. The map simplified the rating process and helped to choose the reason for dependency when creating the activity relationship profile. Both concepts give a base for how activities should be organised and which activities that directly depend on each other. The VSM is anyhow recommended for this purpose because it is easy to understand and it gives a detailed picture.

In figure 44 it is seen that both VSM and Benchmarking cover all of the ware-house operations, but with Benchmarking it is possible to define goals and identify defects within each operation. With the VSM it is possible to define improvement areas between and within each operation. Both with Activity Profiling and VSM waiting time could be reduced. Waiting time could reduced by the calendar-clock profile which facilitates planning and helps the warehouse to plan for personnel

87

and equipment. Waiting orders are reduced by the VSM by eliminating unneces-sary starts and stops in the material stops.

One complication found with the framework is that the framework is time-consuming to apply. It takes time to apply all three concepts.

RQ3: Why should VSM be included in the framework and what benefits were found by applying this concept?

Both the Activity Profiling and Benchmarking are established methods for evalu-ating a warehouse. Activity Profiling provides comprehensive information about how the activities involved in the warehouse operate and how the customers and suppliers impact on the warehouse. Benchmarking provides measurable goals and measurements for assessment. Both of these concepts contribute with much valua-ble information in order to evaluate the warehouse, but none of them can provide any visualisation of the entire material flow.

When identifying the current state, VSM plays a crucial part. It identifies how material moves around at the site: from arriving to departure, between the ware-houses and the tents, back and forth at the site and within the warehouses. The advantage with VSM is that it can identify the entire material flow in a simple map, which will be essential when discussing areas of improvement. Furthermore, it provides a snapshot of the actuality involving waiting time, storage areas along the flow, personnel involved in different activities, the equipment needed and time required at different activities.

One essential part of VSM is to identify wastes along the flow, since those are not adding any customer value. As described in the answers for RQ1 and RQ2, Activi-ty Profiling and Benchmarking contribute both to the identification of wastes and are in some cases even preferred over VSM for identifying a specific waste. To think about wastes however originates from VSM. The VSM has been essential for the identification of areas of improvement.

To motivate the choice of VSM in this framework it is also said in the frame of references that warehouses with high operation efficiency reduce their work con-tent by eliminating material handling steps and by minimising travel time, which is done with VSM. Methods are needed to streamline the flow of material, simpli-fy the decision-making procedure and eliminate non-value adding activities in a supply chain. These statements do perfectly match the purpose of a VSM; to sim-plify decision making, eliminate non-value adding activities and streamline flow of material and information. A warehouse is an important aspect of modern supply chains, which also motivates the use of VSM in this framework.

88

8.2 Future research Some fields of interest are not included in the study, due to the time limit of the project. They are anyhow presented and discussed to highlight potential future research. For this study, the research field has been limited to the warehouses and the boundaries of the warehouses. One interesting topic for a future research is whether this framework would be applicable for the entire supply chain in order to investigate where value is added to the product and where wastes occur.

As discussed in the analysis of the framework, not all of the profiles constituting the Activity Profiling were applied in this research. One profile that would be of interest to apply and to investigate how it would affect the results of the other pro-files and the other concepts is the calendar-clock profile. This profile takes into account how the customer demand depends on seasonality. This is not relevant for the company in this case, but it would be interesting in a future research to investi-gate how this impacts the result and how applicable the framework is for that situ-ation. Also the investment profile and the inventory profile would be interesting to apply in future use of this framework.

This study has been limited to map the physical material flow, partly because of the time limit of the project. When mapping a VSM, as it is supposed to be done, the information flow is included as well. Therefore, it would be interesting to in-vestigate how the framework could be expanded to include the information flow. Especially since Activity Profiling normally includes the information flow when performing the different profiles. This area could contribute to interesting findings, since the information flow plays a significant part of any warehouse organisation.

89

References Literature: Abdi, F., Shavarini, S.K. & Seyed Hoseini, S.M. (2006) Glean Lean: How to use Lean Approach in Service Industries, Journal of Services Research, vol. 6, Special Issue, July, pp. 191-206 Alfa Laval’s Homepage, www.alfalaval.com (2014-02-06) Arbnor, I, & Bjerke, B. (2009) Methodology for creating business knowledge, 3rd edition, SAGE Publications, Inc. Balk, B.M. & de Koster, M.B.M. (2008) Benchmarking and Monitoring International Warehouse Operations in Europe, Production and Operations Management Society, vol. 17, no. 2, March–April, pp. 175–183 Bartholdi, J.J. & Hackman, S.T. (2010) Warehouse & Distribution Science, Release 0.93, The supply Chain and Logistics Institute; School od Industrial and Systems Engineering; Georgia Institute of Technology Bartholomew, D. (2008) Putting Lean Principles in the Warehouse, Lean Enterprise Insti-tute, lean.org Bicheno, J. (2009) Ny verktygslåda för Lean - för snabbt och flexibelt flöde, Revere AB Bozer, A.Y. (2012) Warehousing in the Global Supply Chain: Preface, London: Springe-Verlag Christopher, M. & Towill, D. (2001) An integrated model for the design of agile supply chains, International Journal of Physical Distribution& Logistics Management, vol. 31, no.4, pp. 235-246 Constantino, N., Dotoli, M., Falagario, M., Fanti, M.P. & Mangini, A.M. (2012) A model for supply management of agile manufacturing supply chains, International Journal of Production Economics, vol. 135, no. 1, January, pp. 451-457 de Koster, R., Le-Duc, T. & Roodbergen, K.J. (2007) Design and control of warehouse order picking: A literature review, European Journal of Operational Research, vol. 182, no. 2, October, pp. 481-501 Denscombe, M. (2009) Forskningshandboken - för småskaliga forskningsprojekt inom samhällsvetenskaperna, 2ed edition, Lund: Studentlitteratur AB Dharmapriya, U.S.S & Kulatunga, A.K. (2011) New Strategy for Warehouse Optimization - Lean Warehousing, Proceeding of the 2011 International Conference on Industrial Engi-neering and Operations Management, Kuala Lumpur, Malaysia, January 22-24, pp. 513-519 Dotoli M., Petruzzelli G. & Turchiano, B. (2012) A Lean Warehousing Approach Using Unified Modelling Language and Value Stream Mapping: A Case Study, 9th International Conference of Modeling, Optimization and Simulation, June 06-08, Bordeaux – France Dotoli, M., Epicoco, N., Falagario, M. & Constantino, N. (2013) A Lean Warehousing Integrated Approach: A Case Study, Emerging Technologies & Factory Automation (ET-FA), 2013 IEEE 18th Conference, September 10-13, pp. 1-7

90

Eisenhardt, K.M. (1989) Building Theories from Case Study Research, Academy of Man-agement Review, vol. 14, no. 4, pp. 532-550 Frazelle, E. (2002a) World-Class Warehousing and Material Handling, McGraw-Hill Companies, Inc. Frazelle, E. (2002b) Supply Chain Strategy, McGraw-Hill Companies, Inc. Garcia, F.C. (2003) Applying Lean Concepts in a Warehouse Operation, lean-automation.com Gu, J., Goetschalckx, M. & McGinnis, L.F. (2010), Research on warehouse design and performance evaluation: A comprehensive review, European Journal of Operational Re-search, vol. 203, no. 3, June, pp. 539-549 Gunasekaran, A., Patel, C. & Tirtiroglu, E. (2001) Performance measures and metrics in a supply chain environment, International Journal of Operations & Production Management, vol. 21, no. 1/2, pp. 71-87. Hackman, S.T., Frazelle, E.H., Griffin, P.M, Griffin, S.O. & Vlasta, D.A. (2001) Bench-marking Warehousing and Distribution Operations: An Input-Output Approach, Journal of Productivity Analysis, vol. 16, no. 1, July, pp. 79-100 Hines, P. & Rich, N. (1997) The seven value stream mapping tools, International Journal of Operations & Production Management, vol. 17, no. 1, pp. 46-64 Innovation and Best Practice for Business Success, Logistics & Operation, http://www.koganpage.com/news/warehouse-key-performance-indicators-whats-relevant?returnUrl=http://www.koganpage.com/browse/logistics-operations (2014-0-07) Kovács, G. & Spens, K.M. (2005) Abductive reasoning in logistics research, International Journal of Physical Distribution & Logistics Management, vol. 35, no. 2, pp. 132-144 Lasa, I.S., Laburu, C.O. & De Castro Vila, R. (2008) An evaluation of the value stream mapping tool, Business Process Management Journal, vol. 14, no. 1, pp. 39-52 Lewczuk, K. & Zak, J. (2013) Supportive Role of Dynamic Order Picking Area in Distri-bution Warehouse, Journal of Traffic and Logistics Engineering, vol. 1, no.1, pp. 37-40 Liker, J.K. & Meier, D. (2006) The Toyota Way Fieldbook – A Practical Guide for imple-menting Toyota’s 4Ps [Electronic], The McGraw-Hill Companies, Inc. Manzini, R. (2012) Warehousing in the Global Supply Chain, London: Springer-Verlag Morse, J.M. (1999) Qualitative Generalizability, Qualitative Health Research, vol. 9, no. 1, January, pp. 5-6 New, C. (1993) The Use of Throughput Efficiency as a Key Performance Measure for the New Manufacturing Era, The International Journal of Logistics Management, vol. 4, no. 2, pp. 95-104 Okeudo, G. N & Uche, C. D. (2013) Correlational Analysis of Distribution Cost and Freight Characteristics of Manufactured Goods (Case Study of Unilever Nigeria Plc), Journal of Logistics Management, vol. 2, no 1, pp. 26-34 Park, B.C. (2012) “Order Picking: Issues, Systems and Models”, in Manzini, R. (ed.) Warehousing in the Global Supply Chain, London: Springer-Verlag Rimiené, K. (2011) Supply chain agility concept evolution, Economics & Management, vol. 16, pp. 892-899

91

Roberts, P., Priest, H. & Traynor, M. (2006) Reliability and Validity in Research, Nursing Standard, vol. 20, no. 44, April, pp. 41-45 Rother, M. and Shook, J. (2004) Lära sig se – Att kartlägga och förbättra värdeflöden för att skapa mervärden och eliminera slöseri, version 1.2, Edita Norstedts Tryckeri Saunders, M., Lewis, P. & Thornhill, A. (2003) Research Methods for Business Students, 3rd edition, Harlow: Pearson Education Limited Serrano, I., Ochoa, C. & de Castro, R. (2008) Evaluation of value stream mapping in man-ufacturing system redesign, International Journal of Production Research, vol. 47, no. 16, August, pp. 4409-4430 Stevens, G.C. (1989) Integrating the Supply Chain, International Journal of Physical Dis-tribution & Logistics Management, vol. 19, no. 8, pp. 3-8 So, K.C. & Pinault, S.C. (1988) Allocating buffer storages in a pull system, International Journal of Production Research, vol. 26, no. 12, pp. 1959-1980 Suzaki, K. (1987) The new manufacturing challenge - techniques for continuous improve-ments: Eliminating Waste, The Free Press Thurmond, V.A. (2001) The Point of Triangulation, Journal of Nursing Scholarship, vol. 33, no. 3, March, pp. 253-258 van den Berg, J.P. (1999) A literature survey on planning and control of warehousing systems, IIE Transactions, vol. 31, pp. 751-762 Voss, C., Tsikriktsis, N. & Frohlich, M. (2002) Case research in operations management, International Journal of Operations & Production Management, vol. 22, no. 2, pp. 159-219 Yin, R.K. (2009) Case Study Research, Design and methods, 4th edition, SAGE Publica-tions, Inc. Interviews: Participant observations in the warehouses and at the yard: 2014-01-09 – 2014-01-17 Workshop with warehouse A: Team Manager and Unit Manager, 2014-02-18 Workshop with warehouse B: Team Manager, Unit Manager and Team Leader for Receiv-ing and Shipping, 2014-02-12 VSM with warehouse A: Team Manager, Team Leader for Manufacturing Frames Com-ponent, Team Leader for Receiving and Shipping, 2014-02-20 VSM with warehouse B: Team Manager, Team Leader for Receiving and Shipping, 2014-02-25 Visits to the warehouses: 2014-06-06

92

Appendix A

Interview guide:

Questions asked during the workshop:

• How many lines do you pick each year? • How many hours is a workday? • How many workers are working in the warehouse? • What is value for your customer? • What is value for the warehouse? • Are the majority of picks broken case, full cases or pallets? • What do you define as problem areas in the warehouse? The participant

wrote the problem areas at post-its and afterwards they were discussed.

Questions asked during the VSM:

First question:

• Which are the largest customers?

Asked for each storage location:

• For how long is a product stored here? • How many products are stored here at the moment? • How large is the storage area? • How many products could be stored here? • What kind of products is here stored? • What do the products come from? Alternative, what is the step before this

step?

Asked for each activity:

• What kind of products is handled at this station? • Could everyone work at this station or do you need any special training? • Do you need any special equipment to handle this activity? • How many hours a day do you spend at this activity? • What do the products come from? Alternative, what is the step before this

step?

93

Asked for each transport:

• What is the distance? • How many transports each day?